First Citizens BancShares, Inc. (NASDAQ: FCNCA) — Pay Book Value to Ride Shotgun With America’s Best Bank-Deal Family

Date: June 12, 2026 Price reference: ~$2,115 (Class A, FCNCA, ≈ June 11, 2026) · Total economic shares ≈ 11.6M (Class A 10.58M + Class B 1.01M) · Market cap ≈ $24.5B Sector: Financials — U.S. Super-Regional Bank Holding Company · CIK: 0000798941 · HQ: Raleigh, North Carolina

Standing disclaimer: This is independent research for general information only. The analytical body (Sections 1–15) carries no buy/sell recommendation and no price target — valuation is discussed only as embedded expectations and scenarios. The sole, deliberate exception is the clearly-labeled “Claude’s Take” block immediately below, which is the author’s own opinion and not investment advice.

⚡ Claude’s Take

This block is the author’s own subjective opinion. It is general information, not investment advice. The analytical body (Sections 1–15) below it takes no position and contains no price target.

Verdict: BUY-on-weakness / HOLD here — a fortress-balance-sheet, owner-operated franchise run by the best opportunistic bank-deal allocators in America, at a fair-not-cheap ~1.25x tangible book. Accumulate toward ~$1,800 (≈1.05–1.10x TBV). Not a short at any sane price. Directional zone: Fair value ~$1,950–2,400 (≈1.15–1.40x forward tangible book on a normalized ~10.5–11.5% ROTCE compounding into low-double-digit TBVPS growth). The current ~$2,115 sits in the lower-middle of that band. The bull case (~$2,500–2,800) needs the family to land one more bargain acquisition or ex-accretion NIM to stabilize while the buyback keeps shrinking the float; the bear case (~$1,500–1,750) is ex-accretion margin eroding toward the high-2%s with ROTCE drifting under 9%.

Tag: “You’re not buying a moat — you’re renting one of the great capital allocators in banking, at book value.”

First Citizens is a paradox the market keeps mispricing in both directions. On the raw economics it is a below-average super-regional: FY2025 ROA of 0.96%, ROTCE of ~10–11%, and a ~63% efficiency ratio all sit beneath the peer median (Fifth Third earns ~17% ROTCE at ~57% efficiency). A purist looks at those numbers, notes there is no firm-wide moat, and walks away. But that misses the entire point of the stock. What you are actually underwriting is a century-old, family-controlled capital-allocation machine that bought CIT all-stock and Silicon Valley Bank out of FDIC receivership at the absolute bottom — booking two bargain-purchase gains (~$0.4B and ~$9.8B) and zero goodwill — and that pays its CEO on growth in tangible book value per share plus dividends, not on asset size or gross EPS. Tangible book per share has roughly tripled in four years. The CEO, Frank B. Holding Jr., put ~$2.3M of his own money into discretionary, non-10b5-1 open-market purchases in June 2026 at ~$1,850 — the rare insider signal that means what it says. You can’t check this management (the Holding family controls ~60% of the vote through 16-vote Class B stock), so you have to trust them — and the 100-year record says you can.

The reason I land on accumulate-on-weakness rather than table-pounding buy is honest: the reported-earnings tailwind has peaked. Purchase-accounting accretion that inflated NIM to 3.92% is rolling off fast (NIM now 3.09% and management guides FY26 net interest income down), core ex-accretion NIM is also compressing, and the bank deliberately spent its SVB capital windfall down (CET1 13.0%→10.8%) via a ~$5.7B buyback that retired >20% of the shares. The fortress is now merely strong, not bulletproof, just as the bank approaches the $250B Category-III regulatory cliff (which would force the ~$1.2B HTM securities mark into capital) and a ~$30B FDIC note maturity in March 2028. At ~1.25x TBV the price already pays for a continuation of ~11% ROTCE; you are not stealing it. Framing: quality-compounder-with-a-capital-allocation-moat at a fair price — a “trust-the-jockey” value-with-optionality holding, not a deep-value falling knife.

Conviction: medium-high. Flips decisively bullish if the family lands another counter-cyclical bargain deal, or ex-accretion NIM stabilizes in the low-3%s with ROTCE holding ~11%+ while the buyback continues — at which point 1.4x TBV is reachable. Flips bearish if ex-accretion NIM keeps grinding toward the high-2%s and ROTCE slips below 9%, a venture-credit or office-CRE shock hits the SVB/CRE books, the Direct Bank’s hot $45B of deposits run in a stress, or — the one thing that would break the thesis — the family does a value-destructive large acquisition at a premium to book instead of a bargain.

1. Executive Summary

First Citizens BancShares is the ~$230B-asset holding company for First-Citizens Bank & Trust Company, a Raleigh, North Carolina institution founded in 1898 and controlled for three generations by the Holding family through super-voting Class B stock. In the span of fifteen months it transformed itself from a sleepy ~$60B Southeastern retail bank into a top-20 U.S. bank through two of the best-priced large bank transactions of the cycle: the all-stock acquisition of CIT Group (closed January 3, 2022), which added national middle-market commercial lending, factoring/asset-based finance, and a railcar-leasing business; and the FDIC-assisted acquisition of Silicon Valley Bank (closed March 27, 2023, at the depth of the regional-banking panic), which added the country’s premier technology/innovation and venture-banking franchise. The SVB deal alone produced a ~$9.8 billion non-taxable bargain-purchase gain, driving 2023 net income to an artificial $11.5 billion that must be normalized out of every trend and multiple.

On its underlying economics, FCNCA is a well-run but structurally ordinary spread lender — with one extraordinary feature: its owners. The base business earns sub-peer returns (FY2025 ROA 0.96%, ROTCE ~10–11%, efficiency ~63% — all below the super-regional median), partly because management deliberately over-capitalizes the balance sheet and partly because ~27% of its deposits are higher-cost, rate-chasing “Direct Bank” online money that blunts any funding advantage. There is no broad, firm-wide moat. What there is — and what the entire bull thesis rests on — is a capital-allocation and cultural advantage: a patient, founder-family allocator that buys distressed assets below fair value, refuses to pay up at the top, returns excess capital aggressively, and ties executive pay to per-share tangible-book growth. The financial fingerprint of that discipline is unmistakable: tangible book value per share roughly tripled in four years, with zero goodwill created.

The defining tension of the next two years is the fade of the windfall. Reported profitability has peaked for this cycle. Purchase-accounting accretion — the rolled-in discount on the CIT and SVB loan books — contributed ~$740M to net interest income in 2023 but only ~$251M in 2025 and is now running at ~$156M annualized; it inflated reported NIM by ~42bp at the peak and ~8bp today. Net interest margin has compressed from 3.92% (2023) to 3.25% (2025) to 3.09% (Q1 2026), and management guides 2026 net interest income to decline to $6.5–6.8B with NIM troughing around Q3 2026. Simultaneously, the company spent its SVB capital cushion: CET1 fell from 13.4% (2023) to 10.8% (Q1 2026) as it repurchased ~$5.7B of stock (>20% of shares), terminated the FDIC commercial shared-loss agreement (adding ~$8.8B of risk-weighted assets), and began voluntarily prepaying the $36B / 3.50% FDIC purchase-money note ahead of its March 2028 maturity. It is doing all of this as it approaches the $250B Category-III threshold, which would force the ~$1.2B held-to-maturity securities mark into regulatory capital and impose LCR/NSFR and full stress-testing.

Financial quality is solid and conservative, not spectacular. Credit is benign (allowance/loans 1.05%, net charge-offs 0.45%, nonaccruals rising modestly to 0.96%), with the riskiest pockets — the residual SVB “investor-dependent” venture book (6.1% reserve) and general office CRE (1.3% of loans, ~8% reserve) — small and heavily reserved. The securities book is short-duration and the HTM unrealized loss (~$1.2B, ~6% of tangible common equity) is far less threatening than at most peers — a genuine relative strength. Operating cash flow exceeds net income; there is no goodwill to impair; and there is no aggressive non-GAAP storytelling (the company discloses almost no adjusted metrics at all).

Valuation is fair, not cheap, on absolute economics — but cheap versus the company’s own history. At ~$2,115 the stock trades at ~1.25x tangible book, ~1.22x common book, and ~12x TTM earnings — roughly in line with peer P/E and at a justified premium to book for a ~11% ROTCE bank. On the company’s own ten-year valuation history it screens inexpensive (composite 29th percentile; price-to-sales at the 14th). A justified-P/TBV framework (ROTCE ~11%, cost of equity ~10%, growth ~4%) supports ~1.15–1.30x, so the price embeds roughly “ROTCE holds ~11% with mid-single-digit book growth” — undemanding, with the family’s M&A optionality thrown in for free. The institutional body below takes no position; Section 10 and Section 11 frame the embedded expectations and the variant perception.

Bottom line for the committee: a fortress-balance-sheet, owner-operated bank whose base returns are sub-peer but whose capital allocation is genuinely top-tier, trading at a fair price near tangible book with the reported-earnings tailwind fading and a regulatory cost step-up ahead. The verdict on the next leg depends almost entirely on whether you believe the family keeps doing what it has done for a century.

2. Business Overview

What the company does. First Citizens BancShares, Inc. is the financial holding company for First-Citizens Bank & Trust Company (FCB). It is a diversified, deposit-funded commercial and consumer bank that makes money three ways: (1) net interest income — the spread between yields on loans, leases and securities and the cost of deposits and borrowings, which is ~70% of revenue; (2) noninterest (fee) income — wealth management, card and treasury/payments, factoring and capital-markets fees, plus operating-lease rental income from the railcar fleet; and (3) increasingly, scale-and-skill in acquisition integration, converting distressed franchises into earning assets at a discount. (FACT — FY2025 10-K, filed 2026-02-24.)

Scale and footprint. At year-end 2025 FCNCA held $229.7B in assets, $161.6B in deposits, $147.9B in loans and leases, and $22.2B in total equity, with ~17,900 employees and ~500 branches concentrated in the Southeast (NC ~209, SC ~119, CA ~72) plus a Mid-Atlantic, Midwest and Western presence and a nationwide online “Direct Bank.” (FACT — FY2025 10-K.) That makes it the #3 bank by deposit share in North Carolina and #4 in South Carolina, and a top-20 U.S. bank by assets.

Segments. A material FY2025 reporting change: the company collapsed from four reportable segments to three, folding the former standalone SVB Commercial segment into Commercial Bank (2023–24 recast). The current segments (FY2025) are:

(FACT — FY2025 10-K segment footnote.)

Loan book. Period-end loans of $147.9B are ~81% commercial / 19% consumer. Within commercial: C&I ~30% ($44.7B); capital-call lines ~21% ($31.8B, +25% YoY) — the SVB Global Fund Banking engine and the single largest growth driver; owner-occupied CRE ~12%; non-owner CRE ~16% ($23.8B); and the residual SVB “investor-dependent” early/growth-stage venture book at just ~2% ($2.8B, −13% YoY). Consumer is overwhelmingly residential mortgage (~15%). (FACT — FY2025 10-K, loan tables.)

Deposit franchise. $161.6B of deposits, of which only ~25% are noninterest-bearing (down from 27% in 2023) and the interest-bearing book costs ~2.98%. Roughly $44.8B (~28%) is the high-rate Direct Bank (online savings ~3.68%), with the balance in branch checking, money-market and time deposits. Uninsured deposits are ~38%. (FACT — FY2025 10-K, deposit yield tables.)

Revenue mix and recurrence. Net interest income dominates (~$6.8B of ~$9.5B of revenue in 2025); noninterest income (~$2.7B) is a mix of recurring wealth/card/service fees and the recurring-but-cyclical Rail rental income, with lumpier factoring and capital-markets fees. The business is therefore mostly recurring spread-and-fee income, with the cyclicality concentrated in Rail leasing and the venture/innovation book.

Verdict: A genuinely diversified, three-business bank — a Southeastern retail/wealth franchise, a national commercial/specialty-finance/innovation bank, and a niche railcar lessor — funded by a deposit base that is only modestly low-cost because roughly a quarter of it is rate-chasing online money. The composition is more interesting than the average regional; the economics, as we show, are not.

3. Industry Dynamics

Structure. U.S. regional and super-regional banking is a structurally mediocre industry: products are near-fungible (a deposit and a C&I loan are commodities), competitive intensity is high, the business is capital-intensive and cyclical, and it is among the most heavily regulated sectors in the economy. Profit pools are dominated by net interest income, where individual banks are price-takers on both sides of the balance sheet, and by fee businesses (payments, wealth, capital markets) where differentiation is possible but contested. Switching costs, historically the industry’s best friend (consumer inertia, embedded treasury-management relationships), are eroding as digital account-opening, rate-comparison tools and fintech alternatives lower the friction of moving money — a structural negative the 2023 episode made vivid.

The 2023 regional-bank crisis reshaped the landscape — in FCNCA’s favor. The failures of Silicon Valley Bank, Signature and First Republic proved that “sticky” deposits can flee at internet speed when confidence cracks, permanently raising the risk premium on uninsured, concentrated, or rate-sensitive funding. But the same episode created the forced-seller environment in which a disciplined, over-capitalized consolidator thrives. FCNCA was the buyer of SVB out of receivership; JPMorgan took First Republic; the weak exited and the survivors consolidated. In Marathon-capital-cycle terms, the regional sector is in the bust-to-early-recovery phase — capital and deposits fled, supply contracted, and the remaining disciplined players are taking share. FCNCA, as a serial bargain-buyer (CIT, SVB, and now a 138-branch carve-out from BMO), sits on the favorable side of that cycle.

Regulatory tiering is the defining FCNCA-specific industry dynamic. Under the tailoring framework, FCNCA is currently a Category IV organization ($100–250B in assets): it faces enhanced liquidity-risk management, company-run liquidity stress testing, and resolution planning, but is exempt from the strictest capital and liquidity rules and runs its first supervisory CCAR stress test only in 2026. At $229.7B and growing — with the BMO deal and organic growth pushing it toward $250B — it is approaching the Category III cliff. Crossing $250B would (a) force accumulated other comprehensive income, including the AFS securities mark, into CET1 (today FCNCA opts out, so its rate marks bypass capital); (b) impose the Liquidity Coverage Ratio and Net Stable Funding Ratio (at the reduced 85% calibration for non-GSIBs); © bring full supervisory stress testing; and (d) likely a long-term-debt requirement. This is a real, quantifiable cost and capital headwind that the bull case must underwrite.

Rate and deposit-beta backdrop. As policy rates eased, FCNCA’s interest-bearing deposit cost fell ~49bp to 2.98% in 2025, but its deposit beta is structurally elevated because roughly half of interest-bearing deposits (the Direct Bank savings and money-market book) are rate-sensitive. In a falling-rate world, asset yields reprice down faster than this hot funding, compressing core margin — which is exactly what the NIM trajectory shows.

Verdict: structurally unattractive industry — commoditized, cyclical, capital-intensive, heavily regulated, with eroding switching costs — but FCNCA is favorably positioned within it as a well-capitalized consolidator on the recovery side of the capital cycle, with a regulatory cost step-up looming at the $250B line.

4. Competitive Position

The honest answer: FCNCA is a very-well-run, conservatively-capitalized, family-stewarded bank with pockets of genuine local scale and one specialty niche — but its blended economics are those of an ordinary-to-slightly-above-ordinary spread lender, not a wide-moat compounder. The tell is the financial outcome. A real, broad moat shows up as superior through-cycle returns on capital; FCNCA’s do not.

(FCNCA: FACT, FY2025 10-K. Peer figures directional, from public reporting.)

Now the moat analysis, applying the Greenwald taxonomy and being deliberately skeptical:

(a) The legacy NC/SC retail deposit base — a real but localized advantage. As the #3/#4 deposit bank in the Carolinas with a ~100-year brand, FCNCA has the closest thing to a Greenwald local economies-of-scale plus customer-captivity moat in the company: dense branch coverage, multi-generational relationships, and the habit/switching-cost stickiness of primary-checking households. The financial corroboration is that General Bank earns the best segment economics (net income +25% in 2025, falling provision). But it is narrow — two states — and the blended deposit cost (2.98% interest-bearing, 25% NIB) is not the cheap-funding signature of a true low-cost-deposit franchise, because the Direct Bank dilutes it. A real local moat, too small to drive a $230B balance sheet.

(b) The SVB / venture-banking franchise — recovering, valuable, but contested. The 2023 narrative was that FCNCA had bought an impaired, eroding asset. The data refute that: SVB Commercial deposits grew (≈$34.7B in 2023 to ≈$36.6B in 2024); the lending engine — Global Fund Banking capital-call lines — grew +25% to $31.8B in 2025 with zero charge-offs across three years; and the risky early-stage “investor-dependent” book was deliberately shrunk to ~2% of loans. The franchise stabilized, de-risked and is now a growth driver. Its moat character is relationship plus switching-cost plus specialized-underwriting — few banks can underwrite capital-call facilities or early-stage venture credit at scale. But it is contested: JPMorgan (via First Republic), HSBC (which poached SVB teams), Stifel and others compete hard, and 2023 proved the “network” of venture relationships follows talent and can move at internet speed. A recovering, valuable specialty-finance niche — not a durable network-effect moat that suspends mean reversion.

© Rail leasing — a capital-cycle commodity with a specialty veneer. 128,400 railcars, ~96% utilization, and Q4 lease re-pricing at ~117% of expiring rates look impressive — but that is the up-leg of a classic Marathon capital cycle (railcar supply/demand, steel, commodity volumes). The segment is asset-heavy ($8.9B of operating-lease equipment), capital-intensive, and earns thin returns ($87M net income, ~1% on segment assets). FCNCA is a price-taker on lease rates. A decent counter-cyclical diversifier, no durable advantage.

(d) Family control and long-termism — the strongest cultural advantage, and double-edged. The Holding family’s super-voting control and patient-capital orientation have produced exceptional M&A discipline and opportunism (the CIT and SVB deals were two of the best-priced large bank transactions of the cycle) and underwrite the conservative, over-capitalized balance sheet. This is closest to a Marathon “disciplined capital allocator” edge — about management quality, not barriers to entry. It is double-edged: entrenchment, key-person dependence on Frank Holding Jr., minority-shareholder powerlessness, and the risk of capital held too conservatively (depressing ROE). The single best reason to own the stock — but it is a jockey advantage, not a structural moat.

Greenwald summary. Moat types present: local economies-of-scale + captivity (Carolinas retail) and specialty switching-cost niches (venture/capital-call, factoring, rail) — all sub-scale relative to the whole. There is no firm-wide moat, and the aggregate financial outcome (sub-peer ROTCE) confirms it. What excess return exists is earned by management skill and excess capital — both emulable and mean-reverting — not by an unbreachable barrier.

Verdict: a fortress balance sheet and a top-tier serial consolidator, but a sub-peer earner of returns on capital. You are buying the allocator, not a franchise margin.

5. Growth History and Forward Opportunities

Growth has been overwhelmingly acquired, not organic. Total assets went from ~$59B (2020) to $230B (2025) almost entirely via two deals:

*2023 net income includes the ~$9.8B non-taxable SVB bargain gain — not a run-rate. (FACT — EDGAR XBRL / 10-Ks.)

Organic growth is modest. Total loans grew ~6% in 2025, but General Bank loans were roughly flat; the growth came from the SVB-related Global Fund Banking capital-call lines (+25%) and select middle-market verticals. Deposits grew ~4%, with gains in the Direct Bank and General Bank.

Forward opportunities:

• SVB-franchise compounding. The clean-up phase looks complete; capital-call lending is the growth engine (management cites record quarterly production and a multi-billion pipeline), the risky early-stage book is shrinking, and innovation deposits are growing. This is the highest-quality organic lever.
• In-market deposit density via the BMO branch deal. Announced October 16, 2025: the acquisition of 138 BMO Bank branches (Midwest/Plains/West) with ~$5.7B of low-cost deposits, ~$1.1B of loans and ~$1B of wealth AUM, at a ~5% net deposit premium (~$285M), expected to close mid-2026. This is a funding-cost play — swapping hot Direct Bank money for granular branch deposits — more than a growth play. Low integration risk (a branch carve-out, not a whole-bank deal).
• Capital return as a per-share growth lever. With ~$2.4–2.8B remaining on the $4.0B 2025 repurchase authorization, the ongoing buyback shrinks the share count and lifts both ROTCE and TBVPS over time — a deliberate burn-down of excess capital toward the new 10.0–10.5% CET1 target.
• Optionality on the next bargain. The family’s demonstrated willingness and ability to act when others are forced sellers is itself a forward “opportunity” that cannot be modeled but has, twice, created enormous value.

The headwind side of growth is that the reported-earnings base is shrinking as accretion fades and NIM troughs; 2026 net interest income is guided down. So forward book-value growth (the metric management is paid on) can continue at a healthy clip even as forward reported earnings stall.

Verdict: high-quality but largely acquired growth, executed with unusual price discipline. The forward story is SVB-franchise compounding, in-market BMO/Southeast density, and burning down excess capital to lift sub-peer returns — credible, management-dependent and capital-cycle-favorable, but not organically high-octane.

6. Financial Quality

Net interest margin and the accretion fade — the dominant quality story. Reported NIM has compressed sharply, and the gap between reported and “core” (ex-purchase-accounting) NIM has nearly closed:

(FACT — FY2025 10-K MD&A; Q1-2026 10-Q.) The accretable yield on the acquired CIT and SVB loan books — the discount being recognized into income as the loans pay down — has fallen from a ~42bp NIM tailwind in 2023 to ~8bp now, running at ~$156M annualized. Critically, core ex-accretion NIM is also compressing (3.50% → 3.01%), so this is not purely an accretion-fade story — underlying margin is under genuine pressure in a falling-rate environment with a rate-sensitive funding base. Management guides FY2026 net interest income to $6.5–6.8B (a low-single-digit decline) with NIM troughing around Q3 2026.

Returns — sub-peer, and the company barely discloses them. A notable quality-of-disclosure point: FCNCA’s 10-K reports only ROA and NIM as headline metrics; it discloses no ROE, no ROTCE, no efficiency ratio, and no adjusted EPS reconciliation in the filing (these appear in the IR earnings supplement, not the 10-K). Computed from the financials:

(INTERPRETATION from FACT inputs; management cited an “adjusted ROE” of ~11.93% for Q4 2025.) Efficiency deteriorated ~470bp in 2025 (revenue down, expenses +6% as the bank invests in the franchise) — a metric to watch against management’s qualitative “mid-50s long-term efficiency” aspiration.

Earnings quality and one-time items. GAAP diluted EPS: $784.51 (2023, bargain-gain-distorted), $189.41 (2024), $165.24 (2025), $42.63 (Q1 2026). Several normalizations matter: (1) 2023 is wholly distorted by the ~$9.8B non-taxable bargain gain (ROA 5.90%, effective tax rate 5.1%) and must be stripped from every trend; (2) acquisition-related expenses remain elevated but declining ($470M in 2023 → $210M in 2024 → $141M in 2025), depressing GAAP with no management add-back provided; (3) a ~$110M allowance release flattered 2025 EPS (provision of $514M ran below net charge-offs of ~$640M) — a modest, non-repeatable tailwind; (4) the FDIC special assessment (FCNCA was a large payer given acquired SVB deposits) hit 2023–24 and is largely out of the 2025 run-rate. Operating cash flow is clean and exceeds net income ($2,923M OCF vs $2,206M NI in 2025), and there is no goodwill to impair (it has held flat at $346M throughout, because the acquisitions generated bargain gains, not goodwill) — both genuine quality positives.

Credit quality — benign, with small, well-reserved hot spots.

(FACT — FY2025 10-K credit tables; Q1-2026 10-Q.) The riskiest pockets are small and heavily reserved: the SVB investor-dependent venture book carries a ~6.1% reserve on just ~2% of loans (and improving), and general office CRE is only ~1.3% of total loans but is stressed (nonaccrual ~9%, reserve ~8%). Capital-call lines — the growth engine — have had zero charge-offs. The trend in nonaccruals (up to 0.96%) bears watching off a low base, but credit is reserved where it matters.

Securities, AOCI and the HTM mark — a relative strength. FCNCA is far less securities-heavy than most regionals (securities ~18% of assets, short duration ~2.7 years). The held-to-maturity book carries a ~$1.2B gross unrealized loss (amortized cost ~$9.6B vs fair value ~$8.5B; ~88% mark) that sits off balance sheet and outside capital — but this is ~6% of tangible common equity, far below the 20–40%-of-CET1 holes at peers like Bank of America or Comerica. AOCI actually swung positive (+$83M) at year-end 2025 as rates eased before reverting modestly negative in Q1 2026. The catch: at the $250B Category-III threshold, AOCI inclusion would expose CET1 to these marks.

Capital. CET1 fell from 13.36% (2023) to 12.99% (2024) to 11.15% (2025) to 10.83% (Q1 2026) — a deliberate ~250bp burn-down driven by the buyback and by the shared-loss-agreement termination (which added ~$8.8B of RWA). Management cut the CET1 target to 10.0–10.5% in Q1 2026. Tier 1 leverage is 9.30%, total risk-based 13.51%. Still well-capitalized — but the extreme 2023 cushion is gone.

Verdict: do economics improve with scale? Only modestly. FCNCA converted two acquisitions into a fortress balance sheet, clean credit, conservative securities positioning and clean cash earnings — but it earns sub-peer returns on that capital, its margin is compressing as accretion fades, and efficiency is deteriorating. High-quality balance sheet, ordinary-to-below-ordinary earnings quality.

7. Capital Allocation

This is the heart of the thesis and, on the evidence, genuinely excellent — exercised inside an entrenched family-control structure that you must trust because you cannot check it.

M&A — counter-cyclical bargain-buying, the strongest single argument for the stock.

• CIT Group (all-stock, closed January 3, 2022): ~6.1M Class A shares issued at an $859.76 close ≈ ~$5.2B of consideration (the headline ~$2.2B figure dates to the October 2020 announcement, before FCNCA’s stock re-rated). It generated a $431M bargain-purchase gain and no goodwill — net assets acquired exceeded the price.
• Silicon Valley Bank (FDIC-assisted, closed March 27, 2023): a ~$9.8B bargain-purchase gain, net of ~$3.4B of taxes (~$13.2B pre-tax), recorded in 2023 noninterest income — buying the premier U.S. innovation bank out of receivership at the depth of the March 2023 panic, again with no goodwill.
• BMO 138-branch deal (announced October 2025): ~$5.7B of deposits at a ~5% premium — buying branches a larger bank is shedding, the same pattern at smaller scale.

In Marathon capital-cycle terms this is a textbook supply-side allocator: deploying capital into distressed/forced-seller situations where assets are available below fair value, booking bargain gains and zero goodwill — the opposite of the goodwill-laden, top-of-cycle bank M&A that destroys value. The caveat the bull must hold honestly: SVB-scale FDIC windfalls are non-repeatable, and integration/credit risk on a now-$230B balance sheet is the offsetting concern.

Buybacks — large, recent, and accretive near book. Repurchases (Class A only) ran $1.24B (2022), zero (2023, correctly paused post-SVB), $1.68B (2024), and $3.03B (2025), with ~$2.4–2.8B remaining on the $4.0B 2025 authorization. Cumulatively the bank has retired >20% of its shares for ~$5.7B since 2024 at average prices of ~$1,900–2,060, i.e. roughly 1.1–1.4x tangible book. At a low-double-digit P/E and near book for a bank earning ~11% ROTCE, these are plausibly value-accretive return-of-excess-capital, not deeply opportunistic repurchases — and the pace is now moderating as CET1 nears target.

Dividends — deliberately minimal. The common dividend grew $2.16 (2022) → $3.89 (2023) → $6.87 (2024) → $7.95 (2025), but on a ~$2,000+ share price that is a ~0.4% yield and a <5% payout of earnings. Capital return is overwhelmingly via buyback. Preferred stock (Series A 5.375%, C 5.625%, D 7.00% issued Nov 2025, E 6.625% issued Feb 2026, plus floating Series B) costs ~$57M+ a year and is being used as a non-dilutive capital-structure swap — issuing Tier-1 preferred to fund common buybacks below book.

Liability management — disciplined. The $36B / 3.50% FDIC purchase-money note (due March 2028) is being voluntarily prepaid (~$2.5B in December 2025, ~$0.5–1.0B/month since), retiring sub-market fixed funding with excess liquidity while keeping optionality. Sensible and low-drama.

Incentive alignment — the standout, with caveats. The CEO’s 2025 total compensation was ~$10.2M on a modest ~$1.04M base, with the bulk in long-term incentives. The primary LTIP metric is the three-year cumulative growth rate of tangible book value per share plus cumulative dividends per share — an explicit per-share value-creation metric, not asset growth, headcount or gross EPS. A bank that pays its CEO on TBVPS+dividends is incentivized to do exactly the accretive, bargain-priced M&A and disciplined buybacks observed — directly rebutting the “growth-for-growth’s-sake” risk. Two caveats: (a) there is no explicit relative-TSR or ROTCE/ROIC hurdle, and TBVPS can be flattered by one-time bargain gains (the SVB windfall mechanically boosted the metric); (b) the >98% say-on-pay vote is largely circular because the family controls ~60% of the vote.

Insider behavior — a genuine conviction signal. In June 2026, Chairman/CEO Frank B. Holding Jr. made discretionary, non-10b5-1 open-market purchases of ~1,250 Class B shares for ~$2.3M at $1,713–$2,070 (transaction code P, aff10b5One = false) — real money at the stock’s all-time-high zone. He is a recurring buyer (also May 2026 and November 2025). Read precisely, though: the pattern is often “sell Class A / buy Class B” — swapping economic exposure for incremental super-voting power, so it is part price-conviction and part deliberate control-consolidation. Director Ellen Alemany (ex-CIT CEO) sold ~$5.2M of Class A — a non-family director cashing out, not thesis-relevant.

Verdict: capital allocation is top-tier — counter-cyclical bargain-buying, per-share-TBV-aligned pay, disciplined buybacks and liability paydown from a fortress balance sheet, and a CEO buying his own stock. The qualifier, addressed in Section 9 and Section 11, is that it all happens inside an entrenched dual-class structure where public holders fund buybacks that perpetually concentrate family control and have essentially no governance recourse.

8. Changes and Headwinds — Last Two Years

Recent-events timeline (selected):

Do the changes strengthen or weaken the thesis? Mixed, and timing-dependent.

• Accretion fade — weakens near-term earnings (the dominant negative). NIM has compressed four straight quarters; FY2026 NII is guided down; ~$100M of accretion rolls off in 2026. Reported profitability has peaked for this cycle; the open question is the level of normalized, ex-accretion earnings power — which is precisely what the analyst cuts are pricing.
• Deliberate CET1 burn-down — strengthens per-share value if the business is durable, but spends the buffer. Returning the SVB windfall via buyback is rational from a position of gross over-capitalization, but it removes the cushion that made FCNCA look bulletproof in 2023 — just as accretion fades and the regulatory regime tightens.
• Shared-loss termination — modestly positive (ends FDIC reporting burden; signals confidence SVB-book losses won’t approach $5B), at the cost of the RWA add.
• BMO deal — modestly positive (cheap, sticky branch deposits diversifying away from hot Direct Bank money; low integration risk).
• SVB-franchise stabilization — positive (capital-call growth, improving venture-book reserve).
• Approaching $250B / Category III — structurally negative (AOCI-in-capital, LCR/NSFR, full stress testing, first CCAR in 2026, likely LTD requirement).

Verdict: the last two years show management harvesting a windfall — spending excess capital, terminating the loss-share, bolting on cheap deposits — while the reported-earnings tailwind reverses. Net effect on the thesis is capital-return-positive, near-term-earnings-negative, with a regulatory cost cliff approaching.

9. Risk Analysis

Catastrophic / total-loss assessment. Low. CET1 of 10.83% (above the firm’s 10.0–10.5% target and the well-capitalized floor), allowance heaviest where credit is riskiest, an HTM mark that is modest relative to $22B of equity and unrealized, and a deposit base that — while 38% uninsured and ~27% hot Direct Bank money — is granular and currently growing. The plausible bad outcome is earnings compression and multiple de-rating (already underway), not impairment of capital. A true total loss would require a simultaneous venture-credit collapse, CRE blowup, and a deposit run exceeding ample liquidity — a low-probability confluence.

10. Valuation Discussion (Embedded Expectations)

Where the stock trades. At ~$2,115 on ~11.6M total economic shares, FCNCA’s equity is worth ~$24.5B. Against ~$1,690 of tangible book value per share (Q1 2026) and ~$1,735 of common book value per share, that is:

On the company’s own ten-year history, the stock is inexpensive: the AZI own-history valuation index places the composite at the 29th percentile, P/E at the 41st, P/B at the 33rd, and price-to-sales at the 14th — i.e. cheaper than ~70–85% of its own post-acquisition history. That history is short and distorted by the windfall, so treat it as a directional tell, not gospel.

Justified-multiple framework. For a bank, the fair price-to-tangible-book follows P/TBV = (ROTCE − g) / (COE − g). Using a normalized ROTCE of ~11%, a cost of equity of ~10%, and sustainable growth of ~4% gives ~1.17x; at a 9.5% COE, ~1.27x. So ~1.25x TBV is roughly fair-to-slightly-rich for an ~11% ROTCE bank — which means the current price embeds little more than “ROTCE holds ~11% with mid-single-digit book growth.” That is an undemanding hurdle, and it leaves the family’s proven M&A optionality essentially unpriced.

Peer cross-check. Fifth Third trades at ~1.97x tangible book on a ~17% ROTCE; FCNCA at ~1.25x on ~11%. The lower multiple is justified by the lower returns — banks are priced on ROTCE — so FCNCA is not “cheap versus FITB” in any naïve sense. On P/E (~12x) it sits squarely in the ~11–13x regional range. The relative-value argument is therefore not “cheaper than peers”; it is “fair on current returns, with free optionality on capital allocation and own-history cheapness.”

Embedded-expectations read. What must be true to justify ~$2,115?

• The market is underwriting correctly: that normalized ROTCE settles around 10–11% (below peers), that accretion-driven earnings step down in 2026, and that the bank is well-capitalized but no longer windfall-rich.
• The market is not paying for: (a) another counter-cyclical bargain acquisition (unmodellable but historically the source of most value created); (b) the ongoing buyback compounding TBVPS as the float shrinks; or © ex-accretion NIM stabilizing rather than continuing to erode. These are the sources of upside.
• The market is exposed if: ex-accretion NIM keeps grinding toward the high-2%s, ROTCE drifts below 9%, the $250B regulatory step-up arrives with the AOCI/HTM mark, or a credit pocket (venture, office) cracks — any of which would justify a re-rate toward ~1.0–1.1x TBV.

Scenario zones (fair-value, for orientation only — no price target):

• Bear (~$1,500–1,750): ex-accretion NIM erodes, ROTCE ~9%, multiple compresses to ~1.0x TBV; a venture/CRE credit scare or deposit wobble.
• Base (~$1,950–2,400): normalized ROTCE ~10.5–11.5%, ~1.15–1.40x TBV, TBVPS compounding mid-to-high single digits via retained earnings + buyback; no transformational deal.
• Bull (~$2,500–2,800): the family lands another bargain deal or ex-accretion NIM stabilizes in the low-3%s with ROTCE holding ~11%+, re-rating toward ~1.4–1.5x TBV.

The current ~$2,115 sits in the lower-middle of the base zone. No price target, no recommendation — the embedded expectation is modest, and the asymmetry comes from the unpriced optionality, not from a low absolute multiple.

11. Variant Perception

Consensus view. The sell side treats FCNCA as a well-managed, fortress-capitalized super-regional whose post-SVB earnings boom is normalizing down — hence the spring-2026 wave of target cuts citing NIM and deposit-pricing pressure. Consensus is roughly “fairly valued, near-term earnings headwind, hold.”

The strongest bull case (the variant worth holding). FCNCA is not a bank you value on this year’s ROTCE — it is a family-controlled compounding machine whose scarce asset is capital-allocation skill. Over a century, and twice in the last four years, the Holding family has deployed capital into forced-seller situations at prices that generated bargain gains and zero goodwill, tripling tangible book per share. They pay themselves on TBVPS growth, return excess capital via buybacks near book, and the CEO is buying his own stock with both hands. At ~1.25x TBV / ~12x earnings — cheap versus its own history — you are paying a fair price for the base business and getting the allocation optionality for free. In a consolidating, forced-seller industry (the favorable side of the capital cycle), that optionality is worth a great deal. Falsification: another deal that destroys rather than creates value, or TBVPS growth stalling for several years.

The strongest bear case. Strip the narrative and you have a sub-peer bank with a fading sugar high. ROTCE is ~10–11% (below peers), efficiency is deteriorating to ~63%, ex-accretion NIM is compressing in a falling-rate world against a hot, rate-sensitive funding base, and the reported-earnings tailwind (accretion) is rolling off. The fortress is being dismantled by buybacks just as the bank approaches a tougher regulatory regime that will pull the HTM mark into capital, and a ~$30B wholesale-funding maturity looms in 2028. Meanwhile public shareholders fund buybacks that entrench a family they cannot vote out, in a stock that already trades above tangible book. Falsification: ex-accretion NIM stabilizes and ROTCE holds ~11%+ through the rate cycle.

The 3–5 assumptions that matter most:

1. Normalized, ex-accretion ROTCE — is it ~11% (fair value supports the price) or drifting to ~9% (re-rate lower)?
2. Will the family keep finding and winning bargain deals, or was SVB a once-in-a-cycle windfall?
3. Deposit stability — does the hot Direct Bank/uninsured base hold in a stress, or is 2023 a template?
4. The $250B regulatory step-up — how much does AOCI-in-capital + LCR/NSFR + LTD cost in returns?
5. Buyback vs. capital buffer — is spending CET1 to 10.0–10.5% prudent harvesting or imprudent at the wrong point in the credit/regulatory cycle?

What would change my mind, in one line each: bullish trigger — a new bargain acquisition or ex-accretion NIM stabilization with ROTCE ~11%+ and continued buyback; bearish trigger — ex-accretion NIM toward the high-2%s with ROTCE <9%, a venture/CRE credit shock, a deposit run, or a value-destructive premium acquisition.

12. Fact vs. Interpretation Table

13. Open Questions

1. Normalized ex-accretion earnings power — what is the true run-rate ROTCE once accretion fully rolls off and NIM troughs (management says Q3 2026)? The single most important unknown.
2. Management’s medium-term ROTCE / efficiency targets — not disclosed in the 10-K; need the IR deck / call detail to anchor the base case.
3. Exact FDIC special-assessment amounts by year (2023–24) and any residual.
4. Average tangible common equity series to compute a precise ROTCE (the 10-K discloses neither ROE nor ROTCE).
5. The next deal — is there a visible pipeline beyond BMO, and would the family stretch on price if forced sellers dry up?
6. Direct Bank deposit behavior in a stress — how price-elastic and how sticky is the ~$45B online book?
7. Category III timing and cost — when does FCNCA cross $250B, and what is the quantified capital/liquidity drag?
8. Rail leasing outlook — where in the railcar capital cycle is the segment, and how durable is the ~117% lease re-pricing?

14. What Must Be True

Bull case — what must be true, and its falsification test.

• Thesis: The family keeps compounding tangible book per share at low-double-digits through disciplined buybacks and the occasional counter-cyclical bargain deal, while ex-accretion ROTCE holds ~11%+; at ~1.25x TBV with free optionality, the stock compounds with book.
• Must be true: (a) ex-accretion NIM stabilizes rather than erodes; (b) credit stays benign through the cycle; © the family continues to allocate capital as well as it has; (d) the $250B regulatory step-up is manageable.
• Falsification test: if, over the next 4–6 quarters, ex-accretion NIM continues to fall and computed ROTCE drifts below ~9% with no offsetting accretive capital action, the compounding thesis is broken and the stock should trade back toward ~1.0x TBV.

Bear case — what must be true, and its falsification test.

• Thesis: FCNCA is a sub-peer bank riding a fading purchase-accounting sugar high, spending its capital cushion via buybacks at the wrong time, into a tougher regulatory regime and a hot, run-prone funding base.
• Must be true: (a) ex-accretion NIM keeps compressing; (b) the regulatory step-up and 2028 funding maturity materially raise the cost of doing business; © a credit or deposit shock exposes the now-thinner buffer.
• Falsification test: if ex-accretion NIM stabilizes in the low-3%s, ROTCE holds ~11%+ through the rate cycle, and deposits remain stable while the buyback continues, the “fading sugar high” bear case is falsified and the stock deserves a re-rating toward ~1.4x TBV.

15. Source Appendix

See the separate Source Appendix (FCNCA_source_appendix.md) for the full citation list. Primary sources: FCNCA FY2021–FY2025 Forms 10-K (filed 2022–2026); Q1–Q3 2025 and Q1 2026 Forms 10-Q; 2026 DEF 14A proxy (filed 2026-03-23); Forms 4 (insider, June 2026 and trailing); selected 8-Ks (BMO branch deal 2025-10-16; shared-loss termination 2025-04-07; preferred issuances; quarterly earnings); SEC EDGAR XBRL financial data; the AZI fundamentals/news feeds; the Q1 2026 and Q4 2025 earnings-call commentary (company IR / public transcript sources, as the AZI transcript feed is stale); and public same-sector peer reporting for framing.

APPENDIX A — Standard Diligence Questionnaire

First Citizens BancShares, Inc. (NASDAQ: FCNCA) — as of June 12, 2026

Supplemental to the research memo. Answers are grounded in the filings, labeled Fact / Interpretation / Assumption where it matters. Where a question does not map to a bank’s business model, the correct sector analog is given.

General

What thoughtful questions have other investors asked about this company? The recurring institutional questions are: (1) What is normalized, ex-purchase-accounting earnings power once the CIT/SVB accretion fully rolls off? (2) Is the SVB venture franchise a durable asset or an eroding one after the 2023 collapse and deposit flight? (3) How much of the post-2023 ROE was a non-repeatable windfall (the ~$9.8B bargain gain, the accretion, the over-capitalization)? (4) How should a minority shareholder think about the dual-class family control and the buyback that perpetually concentrates it? (5) What does crossing $250B (Category III) cost in capital and returns? These map directly to the memo’s Sections 10–14.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: reported earnings are coming off a cyclical/structural high. 2023 was wholly distorted by the ~$9.8B bargain gain; 2024–25 were flattered by elevated purchase-accounting accretion (~$481M then ~$251M in NII) and a benign-credit, high-rate environment. With accretion fading, NIM compressing (3.92%→3.09%), and 2026 NII guided down, reported earnings are normalizing lower. Underlying (ex-accretion) earnings are closer to mid-cycle.

Driven by the external environment or internal actions? Both: the level of recent earnings was driven by external windfalls (FDIC-assisted SVB deal, the rate cycle) and internal skill (deal execution, integration, capital return). The direction now (down) is largely external (falling rates, accretion roll-off).

How stable are revenues? Moderately. ~70% is net interest income (spread, rate-sensitive, currently compressing); the rest is wealth/card/service fees (recurring) plus Rail rental income (recurring but cyclical) and lumpier factoring/capital-markets fees.

Outlook for products/services? How big is the market? Deposits and commercial credit are large, mature, commodity markets growing with nominal GDP. The differentiated pieces — venture/innovation banking and capital-call lending — are niche but structurally growing with the private-capital ecosystem; they are also the most contested and cyclical.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More — digital account-opening and rate transparency are eroding switching costs industry-wide, and the venture-banking niche is now contested by JPMorgan/First Republic, HSBC and others. (Fact/Interpretation.)

How profitable is the business (ROIC/ROE)? Bank analog: ROA 0.96%, ROE ~9.9%, ROTCE ~10–11% (FY2025) — below the super-regional peer median. (Fact/computed.)

How profitable is the industry — competitors, barriers to entry? Industry returns are mediocre and cyclical; barriers are regulatory (charters, capital) rather than economic, and provide stability more than excess returns.

Can the business be easily understood? Mostly — a diversified spread-and-fee bank plus a railcar lessor. The complexity is in purchase-accounting accretion, the dual-class structure, and the FDIC note/shared-loss mechanics.

Can it be undermined by foreign low-cost labor? No — domestic, relationship- and regulation-bound.

Do brands matter? Nature of competition? Switching costs? The First Citizens brand matters locally in the Carolinas (real but narrow captivity). Competition is on rate, service and relationship. Switching costs are moderate for primary-checking/treasury relationships, low for the rate-shopping Direct Bank deposits.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The franchise/deposit value of the Carolinas and SVB businesses is not capitalized (no goodwill was created — the deals booked bargain gains). The Rail fleet carries at depreciated cost below likely market in an up-cycle.

Off-balance-sheet liabilities? The ~$1.2B HTM securities unrealized loss is off-balance-sheet and outside capital today (a real, if modest, economic liability). Unfunded lending commitments (notably capital-call lines) are sizeable but low-loss. The FDIC purchase-money note (~$33B) is on balance sheet.

How conservative is the accounting? Conservative. No goodwill, clean cash earnings (OCF > NI), short-duration securities, minimal non-GAAP storytelling, and credit reserved heaviest where risk is highest. The one tailwind to watch is the modest 2025 allowance release.

How CapEx-hungry is the business? Low for the bank; the Rail segment is genuinely capital-intensive (~$8.9B of operating-lease equipment requiring ongoing fleet investment).

Capital Allocation & Management

How much FCF does it generate and how is it used? Bank analog: ~$2.9B operating cash flow / ~$2.2B net income (2025). Capital is returned overwhelmingly via buyback (~$3.0B in 2025, >20% of shares since 2024) and minimally via dividend (<5% payout, ~0.4% yield), with excess liquidity used to prepay the 3.50% FDIC note.

Significant acquisitions recently? Yes — the thesis-defining ones: CIT (all-stock, 2022, $431M bargain gain), SVB (FDIC-assisted, 2023, ~$9.8B bargain gain), and the pending BMO 138-branch deal (2025, ~$5.7B deposits).

Buying back shares? Issuing to insiders? Aggressively buying back Class A (only); not issuing meaningful new common to insiders. Preferred (Series D/E) issued to fund buybacks non-dilutively. Note: Class-A-only buybacks mechanically concentrate the family’s Class B voting power.

Compensation policy / motivations of management? CEO ~$10.2M (2025) on a modest base; LTIP paid on 3-year TBVPS growth + cumulative dividends per share — strongly per-share-value-aligned (Fact). Caveat: no relative-TSR/ROTCE hurdle, and TBVPS can be flattered by one-time bargain gains. Management’s evident motivation is long-term, family-stewarded compounding of per-share book value.

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — a U.S. C-corp bank holding company; common dividends are ordinary 1099-DIV. Dual class (Class A listed; Class B family-held, 16 votes).

Dividend policy? Deliberately minimal (~0.4% yield, <5% payout); capital return is via buyback.

How profitable is the business? Sub-peer (see above): ROTCE ~10–11%.

Net income vs. cash from operations diverging? No — OCF exceeds NI; the only large historical divergence was the 2023 non-cash bargain gain (correctly absent from OCF). Clean.

Risks & Downside

What would cause the stock to decline? Continued ex-accretion NIM erosion / ROTCE drift below 9%; a venture-credit or office-CRE shock; a deposit run in the hot Direct Bank/uninsured base; the $250B regulatory step-up; or — most thesis-damaging — a value-destructive premium acquisition.

Risk of catastrophic loss? Low. CET1 10.83% (above target and the well-capitalized floor), credit reserved where it matters, the HTM mark modest and unrealized, deposits granular and growing.

Chance of a total loss? Very low. Would require a simultaneous venture-credit collapse, CRE blowup, and deposit run exceeding ample liquidity — a low-probability confluence for a well-capitalized, diversified, conservatively-run bank.

Recent News & Events

Has the business environment changed recently? Yes — falling rates and accretion roll-off are compressing margin (the dominant near-term change); the FDIC shared-loss agreement was terminated (April 2025); the CET1 target was cut to 10.0–10.5% (Q1 2026); analysts cut targets on NIM pressure (spring 2026).

Significant acquisitions? The pending BMO 138-branch deal (announced October 2025, ~$5.7B deposits, close mid-2026).

Change in accounting policies? Segment reporting changed in FY2025 (four segments to three; SVB Commercial folded into Commercial Bank).

Recent changes — new markets, facilities, management? BMO extends the footprint into the Midwest/Plains/West; CRO succession (Rupp → Eklund, June 2026); Diane Morais added to the board (2025).

APPENDIX B — Source Appendix

First Citizens BancShares, Inc. (NASDAQ: FCNCA) — Research Sources

Compiled June 12, 2026. Primary sources prioritized. All SEC filings accessed via EDGAR (CIK 0000798941) and mirrored locally. Figures reconciled to filings; third-party aggregator data (AZI, yfinance) used for orientation and cross-check only, never as primary for a reported figure.

Primary — SEC filings (EDGAR, CIK 0000798941)

1. FY2025 Form 10-K — filed 2026-02-24 (fiscal year ended 2025-12-31). Business/segments, MD&A (NIM, accretion, efficiency, capital), credit tables, securities footnote (HTM/AFS marks), FDIC note & shared-loss disclosures, regulatory (Category IV/tailoring) section, capital actions.
2. FY2024 Form 10-K — filed 2025-02-21. Prior-year comparatives, segment (four-segment) basis.
3. FY2023 Form 10-K — filed 2024-02-23. SVB bargain-purchase gain ($9,808M) and acquisition accounting (Note 2).
4. FY2022 Form 10-K — filed 2023-02-24. CIT acquisition ($431M bargain gain), all-stock consideration.
5. Q1 2026 Form 10-Q — filed 2026-05-08 (period 2026-03-31). Latest NIM (3.09%), CET1 (10.83%), nonaccruals, HTM mark, dual-class share counts, buyback, FDIC-note prepayments.
6. Q1–Q3 2025 Forms 10-Q — filed 2025-05-09 / 2025-08-08 / 2025-11-07.
7. 2026 DEF 14A proxy — filed 2026-03-23. Executive compensation and LTIP metric (TBVPS growth + dividends per share), Holding-family beneficial ownership and Class A/B voting, board independence, related-party transactions, say-on-pay.
8. Forms 4 (insider) — June 2026 (Frank B. Holding Jr. open-market Class B purchases, code P, aff10b5One=false; Ellen Alemany Class A sale) and trailing ~12 months.
9. Selected 8-Ks: BMO 138-branch acquisition (2025-10-16); FDIC Commercial Shared-Loss Agreement termination (2025-04-07); Series D preferred (2025-11-18) and Series E preferred (2026-02-05) issuances; CRO succession (2026-01-14); quarterly earnings releases (2025-01-24 through 2026-04-23); annual-meeting results (2026-05-07).

Primary — earnings calls / management commentary

10. Q1 2026 earnings call (~2026-04-23/24) — NIM trough guidance (Q3 2026), FY2026 NII guidance ($6.5–6.8B), CET1 target cut to 10.0–10.5%, buyback pace, BMO and SVB-franchise commentary. Sourced from company IR / public transcript providers (the AZI transcript feed is stale, ending 2022).
11. Q4 2025 / FY2025 earnings call (~2026-01-23) — adjusted EPS/ROE commentary, FDIC-note prepayment, capital plans.

Secondary — quantitative aggregators (orientation / cross-check only)

12. SEC EDGAR XBRL (via edgar.sh) — multi-year NII, net income, equity, deposits, assets, goodwill, noninterest income, provision, buybacks, dividends.
13. AZI fundamentals feed — snapshot (sector/GICS, employees, description, market cap, short interest, ownership), valuation_index own-history percentiles (composite 29th; P/S 14th).
14. AZI news feed — insider Form 4 rows (June 2026 CEO buys, Alemany sale), analyst-target-reset article (2026-05-23).
15. yfinance (fetch.py) — price ($2,115), market cap, share count cross-check, 52-week range.

Secondary — trade press / analyst notes (qualitative, dated)

16. Banking Dive / PR Newswire — BMO 138-branch deal terms (2025-10-16).
17. Investing.com / Yahoo Finance / Simply Wall St — analyst target resets (JPMorgan, TD Cowen, KBW, Piper Sandler, Deutsche Bank), April–May 2026.

Analytical frameworks

18. Competition Demystified (Greenwald & Kahn) — moat-type taxonomy, ROIC/share-stability tests applied in Section 4.
19. Capital Returns (Marathon / Chancellor) — supply-side capital-cycle analysis applied to the regional-bank consolidation and Rail leasing in Sections 3–7.

Note: third-party sentiment/scores and analyst price targets are treated as signals, not evidence, and never as a basis for a price target or recommendation. Every reported financial figure traces to a filing.