FICompany report
Fifth Third Bancorp FITB
The bet you're really making is that Fifth Third can swallow Comerica, the bank it bought in February for its Texas and California commercial deposits, and come out bigger and more profitable without the deposits walking out or the loans going bad. You're betting the cheap business-checking money Comerica brought over stays put, and that the costs Fifth Third promised to cut actually come out. Right now it is going well, with one thing to watch: last quarter delivered $801 million in net income, up 28% from the prior year's Q2, after the quarter before it got gutted by one-time merger accounting, and the provision for credit losses fell 25% year-over-year. You pay 1.8 times the bank's tangible book value, the middle of where it has traded for a decade and a little above other regional banks.
Key data
FITB · price with moving averages
Source: market data.
The business
Fifth Third is a regional bank built on the Midwest and Southeast, and since February 1 it reaches into Texas and California too, having closed the all-stock purchase of Comerica. It earns money the ordinary way: it takes in deposits, lends most of them out at a higher rate, keeps the spread, and adds fees from wealth management, payments and commercial banking on top. Last quarter net interest income was $2.2B and fees another $1.1B. The prize in a bank like this is cheap, sticky funding, the checking accounts businesses leave sitting there paying little or nothing. Comerica was bought mostly for that commercial deposit base: $24.9B of the liabilities that came over pay no interest at all. The whole bank now runs on $234B of deposits, up from $172B before the deal.
The numbers
The story of the last five quarters is one air pocket, then recovery.
| Quarter | Revenue | Net income | Diluted EPS |
|---|---|---|---|
| Q2 2025 | $3.21B | $628M | $0.88 |
| Q3 2025 | $3.30B | $649M | $0.91 |
| Q4 2025 | $3.28B | $731M | $1.04 |
| Q1 2026 | $3.87B | $165M | $0.15 |
| Q2 2026 | $4.43B | $801M | $0.83 |
The $165M in Q1 was not the business breaking. When a bank buys another and the acquired loans are healthy, it must book a full day-one loss reserve against them through the income statement the moment the deal closes, even though nothing has gone wrong. That charge, plus deal expenses, cratered the quarter. Q2 is closer to the real run rate: revenue up 46% year over year on the enlarged balance sheet, with the merger noise draining out.
Zoom out and the reason the deal exists comes into view.
| Fiscal year | Revenue | Net income | Diluted EPS |
|---|---|---|---|
| 2021 | $7.95B | $2.77B | $3.73 |
| 2022 | $9.08B | $2.45B | $3.35 |
| 2023 | $12.4B | $2.35B | $3.22 |
| 2024 | $13.0B | $2.31B | $3.14 |
| 2025 | $12.9B | $2.52B | $3.54 |
| 2026, 1H to Jun | $8.30B | $966M | $0.98 |
Per-share earnings went backwards for four straight years, $3.73 in 2021 down to $3.14 in 2024, before nudging back to $3.53. That is a bank that had stopped compounding. Comerica is the answer to that stall: consensus has FY2027 earnings at $4.93, a 40% jump over 2025, which is why the stock at 18.5 times trailing earnings sits at only 11 times that 2027 number. The entire investment is whether that $4.93 is real. This memo's view is that it is an execution-and-credit outcome the market is pricing at roughly fair odds, not a giveaway: the print that settles it is cost-save realization and the loss trend over the next two or three quarters.
Management
The CEO pushed through the largest deal in the bank's history, an all-stock purchase that reshaped the balance sheet overnight. Insiders sold $6.7M over the past year against zero buys across 18 sales, plan status not disclosed; the largest were Mitchell Feiger at $4.5M in February and Comerica's Peter Sefzik at $947K in April, the kind of selling acquired executives do as stock vests after a close. Buybacks were switched off in 2026, no repurchases in Q1, to hoard capital for the deal, after $525M in 2025 and $625M in 2024. The record is of a management team that will spend big to grow rather than shrink the share count when it matters.
How it fails or surprises you
Deposit flight. Comerica's value is its cheap commercial deposits, and $24.9B of them pay no interest. If those balances chase higher yields elsewhere now that the deal is done, funding costs climb and the accretion math breaks. The first tell is the noninterest-bearing share and cost of deposits over the next two quarters.
Thin reserves on a bigger book. The Comerica loans came on at fair value with no day-one reserve, so reserves-to-loans dropped from pre-deal levels as the denominator doubled. If charge-offs accelerate into a thinner cushion, provisions snap back hard. Watch the next quarters' net charge-off rate and reserve build against total loans.
The re-rate lands (right tail). If the CEO hits the cost targets and deposits hold, FY2027 earnings of $4.93 put today's price at 11 times, and the market is paying 18.5 times trailing because it does not yet believe it. First reveal: the 2H 2026 efficiency ratio and initial 2027 guidance.
Closing thoughts
This is a name where a specific print settles it. The next two or three quarters of cost-save realization and the charge-off and reserve trend convert that 11-times-2027 number into either genuinely cheap or a value trap, and an ambiguous read (deposits flat, delinquencies drifting up, savings on schedule) leaves you holding a fairly-priced bank while the case proves out. The fatter tail is modestly to the upside if execution holds, because the earnings recovery is mechanical once the merger noise clears, but the left tail is real: a bigger loan book carrying a thinner reserve is exactly how acquirers get surprised a year after the handshake.
The bet is still that Fifth Third can swallow Comerica and come out bigger and more profitable, without the commercial deposits walking out or the loans going bad. It breaks if the cost of deposits climbs while delinquencies rise against a reduced reserve ratio. The pair to watch is the noninterest-bearing deposit share and the net charge-off rate; if both move the wrong way together, the $4.93 does not arrive.
Methodology
The year-to-date row is the sum of the 2 reported quarters of the current fiscal year, diluted EPS included; the five-quarter and five-year tables are the vendor income statements.
Numbers re-derived from as-filed XBRL and the Q2 2026 10-Q (filed 2026-08-04, period 2026-06-30) for CIK 0000035527, over vendor fields.
Q4 2025 has no standalone 10-Q; its net income ($730M) and diluted EPS ($1.03) are derived as the 2025 fiscal year less the filed nine months. FY2026 shown as 1H to June; 1H EPS of $0.98 is the sum of Q1 ($0.15) and Q2 ($0.83).
Insider figures from Form 4 filings over the trailing 12 months; 10b5-1 status not carried in the feed, so plan status not disclosed. Proxy not read this run, so pay design is outside this memo.
Fact check: all quarterly and annual financials verified against filed XBRL within rounding; Comerica deal close date (Feb 1, 2026), deposit totals, and noninterest-bearing liability transfer ($24.9B) verified from 10-Q filing; P/TBV 1.8x and historical range verified from valuation_history; insider trading totals and largest sales verified from Form 4 data; buyback amounts verified from filed cash flow statements. CEO name not independently verified this run (evidence pack did not provide officer roster); credit quality ratios (NCO/loans, NPL/loans, reserves/loans) and specific delinquency bucket metrics not independently verified (FDIC Call Report data not in evidence pack). "Biggest profit in company history" claim removed as unverifiable from available quarterly data. Final analysis verified as of Sep 6, 2026.
Bid Cap
Daily ideas, a 390-name database, and a model long/short book from an investor who mostly covers financials. $70 a month or $700 a year.
Subscribe on Substack


