Column N.A.: the programmable bank behind payments, cards and lending

A bank that sells programmable financial infrastructure

Column National Association combines a federally chartered bank with a technology platform for companies building payments, accounts, cards and lending products. Its customer proposition is direct integration with the regulated bank and its own ledger, rather than a software intermediary connecting a fintech to a separate sponsor. That distinction matters economically: Column can supply both transaction infrastructure and balance-sheet capacity, while remaining responsible for the banking activities behind the software. It is neither an ordinary application vendor nor merely a consumer-facing digital bank. [1, 6]

This profile concerns Column N.A., FDIC certificate 58224 and Federal Reserve identifier 3435948. It does not concern similarly named publishing, public-notice or software businesses. Evidence is current through October 4, 2026; the latest financial period retrieved from the FDIC is June 30, 2026. Subsequent product announcements describe capabilities and company claims, not a September-quarter financial statement. [1, 5]

Charter, ownership and the move from Chico to Salt Lake City

The FDIC identifies Column as a national commercial bank, chartered and primarily supervised by the Office of the Comptroller of the Currency. The institution originated on March 28, 2006 as Northern California National Bank. The OCC’s historical evaluation says Column acquired that bank in June 2021; FDIC history records the legal-name change on August 16, 2021. Column publicly launched its developer-focused business in April 2022. These are different events, not conflicting founding dates. [1, 2, 3, 6]

A material 2026 change is the registered main-office location. FDIC history records a move from 1717 Mangrove Avenue, Suite 100, Chico, California, to 350 East 400 South, Suite 214, Salt Lake City, Utah, effective June 12, 2026 and processed July 16. A Chico branch opening is recorded with the same effective date. The October 2 institution dataset, retrieved October 4, lists Salt Lake City. The move did not change Column into a Utah state-chartered bank: the FDIC continues to show class N and OCC supervision. [1, 2]

Column’s current company page identifies William Hockey as CEO and describes the company as wholly founder- and employee-owned and profitable. That is a management ownership statement, not a published capitalization table. The OCC’s 2023 description said the bank had no holding company at that time; that historical observation is not treated here as proof of the complete ownership structure after the 2026 reorganization recorded by FDIC. [2, 3, 4]

June 2026 balance sheet: rapid growth, modest retained loan share

FDIC-reported assets were $1.768904 billion at June 30, 2026, compared with $833.990 million a year earlier, a calculated 112.1% increase. Deposits rose from $712.443 million to $1.529576 billion, or 114.7%. Gross loans and leases increased 70.4%, from $210.528 million to $358.644 million; net loans after the allowance were $356.632 million. The difference between gross and net loans matters when comparing loan ratios across data providers. [5]

Gross loans represented a calculated 23.45% of deposits. This is a balance-sheet measure, not the proportion of all loans originated through Column that it ultimately retains. Originate-and-sell activity, warehouse financing, payment settlement and customer deposits can make transaction and origination volumes much larger than the stock of loans on a reporting date. A small retained portfolio relative to deposits therefore does not establish a small lending franchise, while a large payments franchise does not establish a large loan book. [5, 7]

Total equity capital was $174.821 million, compared with $85.676 million at June 2025. The calculated equity-to-assets ratio was 9.88%. FDIC reported $183.562 million of Tier 1 capital and a 25.37% total risk-based capital ratio. Accounting equity, regulatory Tier 1 capital and total risk-based capital are distinct measures with different adjustments and denominators; none should be substituted for another in a claim about capital adequacy. These snapshots also do not reveal intraday liquidity requirements or the bank’s confidential supervisory assessment. [5]

Profitability, reported credit quality and what the totals omit

First-half 2026 net income was $54.334 million, up 61.4% from $33.658 million in the first half of 2025. This is year-to-date income, not second-quarter-only earnings. Subtracting the March 2026 cumulative amount of $23.179 million gives calculated second-quarter income of $31.155 million. The FDIC’s annualized year-to-date return on assets was 8.13%; that unusually high reported return should not be read as a forecast of sustainable earnings. [5]

First-half interest income of $68.735 million less interest expense of $18.984 million produced $49.751 million of net interest income. Noninterest income was $77.188 million, or a calculated 60.8% of net interest income plus noninterest income. Noninterest expense was $49.391 million. These totals establish that the bank’s reported economics include substantial revenue outside the conventional lending spread. They do not disclose API pricing, interchange sharing, income by partner, one-off items or the incremental profitability of the products launched in September. [5]

Noncurrent loans and leases were $212,000, equal to 0.059% of gross loans, compared with $935,000 and 0.444% a year earlier. The allowance was $2.012 million. Net charge-offs were negative $425,000 in the first half, meaning recoveries exceeded charge-offs in that reported aggregate. These favorable point-in-time figures apply to loans on the bank’s books. They are not loss rates for every loan Column originated and sold, nor evidence that newer vintages will perform identically. Rapid growth can also change portfolio mix before losses mature. [5]

The product stack and the economics of retaining risk

Column’s original 2022 description emphasized its own core, a direct Federal Reserve connection and developer-oriented lending. The current architecture joins bank accounts with ACH, wires, book transfers and real-time payments. The analytical benefit of a common ledger is fewer reconciliation boundaries between balances and payment instructions. The trade-off is concentration: more of a customer’s operating workflow can depend on one bank and one technology environment. Direct integration changes where operational risk resides; it does not remove it. [6, 14]

The October 2025 credit announcement distinguishes warehouse lines, forward-flow purchases and loan retention. Warehouses finance a partner’s loans before sale or securitization; forward flows commit to buy qualifying loans; retention leaves originated assets at Column. It names Best Egg, Finvari and Parafin and describes different uses of origination and capital. This is evidence of differentiated products, not a universal contract or a disclosed concentration schedule. [7]

The economic difference is fundamental. An origination fee can be earned on a loan sold shortly afterward, but retained credit produces interest income while consuming funding and capital. A warehouse creates exposure to both the financed assets and the borrowing counterparty, subject to collateral and contractual protections. A forward-flow commitment may support a partner’s funding certainty while creating future purchase obligations for the bank. The public product description does not quantify advance rates, recourse, loss-sharing or how much committed capacity is undrawn. Those missing terms limit any attempt to infer risk from a partner logo.

September 2026 expansion: processing, stablecoins and international reach

On September 16, Column announced an in-house issuer processor alongside its existing card issuing and capital capabilities. It described support for Visa and Mastercard programs and functions such as authorization controls, disputes, tokenization and digital wallets. The announcement explicitly allows customers to bring their own processor, capital provider or manufacturer. Consequently, the existence of the new integrated stack does not establish that every existing Column-sponsored card uses Column processing. [8]

The same day, Column described USDC and USDT conversion integrated with its ledger, around-the-clock conversion between supported stablecoins and fiat, and connections to domestic and international payment rails. It named Slash and Brex as collaborators and reported tens of billions of dollars in annualized volume. That is a company-reported run-rate claim with no independently audited period or revenue bridge supplied in the announcement. Blockchain transfers, bank deposits and card obligations remain different legal and operational claims even when presented through one API. [9]

Its accompanying launch covered global banking and multicurrency accounts and payouts. The strategic direction is broader than domestic sponsor banking: Column wants customers to combine accounts, currency conversion, payments and credit within one infrastructure relationship. The inference is a larger potential share of each client’s activity, alongside more complex jurisdiction, sanctions, correspondent and settlement risks. Broad marketing language about global availability is not a country-by-country licensing or product-eligibility determination. [10]

Verified relationships show several distinct bank roles

Mercury’s legal page identifies Column as the provider of banking services including the demand deposit account and debit card and links separate deposit, card and sweep agreements. This is stronger evidence of the bank’s legal role than a logo wall. It also shows why a single fintech relationship can contain several contracts with different account and insurance mechanics. Mercury’s use of Column does not mean all Mercury customers or all balances sit at this one bank. [11]

Regional Management’s March 2026 SEC filing provides a more detailed lending example. Column is lender for secured and unsecured installment products in selected states; Regional performs marketing, processing and servicing. The filing describes reciprocal fees, Column’s continued oversight and control, an initial holding period, potential sales to Regional and continued servicing of retained loans. It also describes liquidity and net-worth covenants. These are actual disclosed program features, not assumptions imported from another sponsor-bank arrangement. [12]

Bilt’s current Card 2.0 overview identifies Column as issuer and Cardless as servicer for the Blue, Obsidian and Palladium products. Issuing a card is distinct from running the rewards business or bearing all costs of its benefits. Bilt’s public overview does not disclose Column’s program-level return, retained balance, loss allocation or full commercial agreement. This profile therefore does not turn Bilt’s growth or card pricing into an estimate of Column earnings. [13]

Column also publicly describes Brex’s account and payments use and Ramp’s corporate-card use in certain international markets. These statements are appropriately narrower than calling Column the exclusive worldwide bank for either company. Multiple bank relationships, contractual migration rights and partner business changes can all affect volumes; the reviewed sources provide no customer-by-customer revenue percentages. [14]

Deposit concentration, fraud and operational resilience

Deposits obtained through platforms can be granular at the end-user level but concentrated at the commercial-relationship level. Thousands of account holders may share one app, one treasury decision-maker or one servicing workflow. That makes partner concentration and depositor concentration different questions. Public balance-sheet totals cannot show how many balances would move together if a large platform changed banks, experienced an outage or lost customer confidence.

The retrieved FDIC dataset returns zero for Column’s estimated uninsured-deposit field. This profile does not interpret that field as proof that every underlying balance is fully insured: a reporting field requires its applicable instructions and account-ownership context. Similarly, reported brokered deposits of zero at June 2026 do not prove that all funding came from branches or that all platform balances are behaviorally stable. Legal deposit classification and economic concentration are separate matters. [5]

Fast payment execution can improve customer experience while shortening the opportunity to stop fraudulent transfers. A common ledger can improve visibility, but permissions, account takeover, screening, reconciliation exceptions and recovery procedures still matter. Stablecoin conversion adds wallet and chain dependencies; card programs add authorization, chargeback and dispute processes. These are risks implied by the disclosed business model, not allegations of identified failures at Column.

The federal agencies’ third-party guidance covers selection, contracting, monitoring and termination. A bank’s decision to own more technology does not eliminate obligations associated with the fintechs distributing its products. Separately, FDIC pass-through coverage depends on ownership and recordkeeping requirements; it is not a blanket guarantee of an application’s availability or every financial product shown in it. [15, 16]

The central analytical question

Column’s evidence-backed differentiator is the combination of a national bank charter, internally built technology, substantial noninterest income and an expanding set of funding products. June financials show strong earnings and rapid balance-sheet growth with low reported noncurrent loans. September announcements broaden the addressable workflow, but their incremental contribution is not yet visible in the financial period examined.

The next comparable bank snapshot is a later quarterly regulatory report, not a marketing traffic counter. A useful comparison would preserve the same legal entity, distinguish cumulative income from quarter-only income, and separate gross loans from net loans. Any apparent step-change also needs to be read against the June relocation and expansion of the product set, without assuming either caused a particular financial result. This methodological discipline matters because the balance sheet can change much faster than the wording on a corporate profile.

The unresolved questions are the durability and concentration of fee income, retained credit exposure by program, deposit behavior during stress, and the cost of operating a wider product set. Aggregate profitability cannot resolve those questions without segment and partner disclosures. The useful distinction is between a verified operating platform and an unverified forecast: the former is supported by bank, partner and regulatory records; the latter would require assumptions not available in the reviewed public record.

Sources

1. FDIC institution directory; retrieved October 4, 2026, index dated October 2Official sourceBack to text: ↑1↑2↑3↑4
2. FDIC structure history, Column certificate 58224; retrieved October 4, 2026Official sourceBack to text: ↑1↑2↑3
3. OCC, Column CRA evaluation dated July 10, 2023; historical 2019–2021 examination periodOfficial source · PDFBack to text: ↑1↑2
4. Column, company overview; checked October 4, 2026SourceBack to text: ↑
5. FDIC BankFind financial data, Column, June 30, 2026 and comparison periods; dollar fields in thousandsOfficial sourceBack to text: ↑1↑2↑3↑4↑5↑6↑7↑8
6. Column, Announcing Column; April 21, 2022SourceBack to text: ↑1↑2↑3
7. Column, Credit at Column; October 3, 2025SourceBack to text: ↑1↑2
8. Column, Full-stack card issuing; September 16, 2026SourceBack to text: ↑
9. Column, Stablecoins; September 16, 2026SourceBack to text: ↑
10. Column, Completing the stack; September 16, 2026SourceBack to text: ↑
11. Mercury, Column policies and agreements; checked October 4, 2026SourceBack to text: ↑
12. Regional Management, Form 8-K filed March 5, 2026, agreement dated March 2Filing / reportBack to text: ↑
13. Bilt, Card 2.0 program overview; updated September 18, 2026SourceBack to text: ↑
14. Column, product and partner overview; checked October 4, 2026SourceBack to text: ↑1↑2
15. Federal banking agencies, third-party risk-management guidance; June 6, 2023Official releaseBack to text: ↑
16. FDIC, pass-through deposit insurance coverageOfficial sourceBack to text: ↑