Each week, Steve is breaking down what’s happening in fintech banking with the kind of clarity you get from someone who’s lived through board debates, pricing standoffs, and product launches that either scaled or crashed. This isn’t surface-level commentary. It’s the real story behind sponsor bank partnerships, embedded finance moves, and BaaS programs that most people only hear about after they’ve already succeeded or failed.

Advertisment

Transform Your Merchant Applications with Under. The Under platform revolutionizes how you handle merchant applications, offering a seamless transition to digital forms. Say goodbye to outdated processes and hello to efficiency. Discover the future of financial applications at https://under.io/ftc 

Advertisement

Ready or Not 2026 is Blasting off

January so far is showing the usual pattern: embedded keeps scaling up, but the moves coming out of D.C. and New York state aren't making it any easier to keep things loose. Headlines feel like they're helping. The fine print is telling sponsors to tighten controls.

Here are the five things that stand out so far:

  • Trump ordered Fannie and Freddie on January 8 to buy $200 billion in mortgage bonds, giving embedded mortgage a short affordability boost. Then he floated a one‑year 10% cap on card APRs starting January 20. If it ever lands, it squeezes card, BNPL, and revolving credit spreads hard in BaaS setups. Sponsors, pricing variability just got more real.

    • Framed as making mortgages and credit cards “more affordable” through lower rates and capped APRs

    • Signals more policy reach into pricing and the need for financial institutions to comply

    • Build a business that can withstand a shifting regulatory environment, variability is the only constant

  • New York DFS finalized nonbank mortgage lenders with 200+ originations in the state yearly now face CRA-style exams on fair access and community service, effective July 7, 2026. Embedded housing platforms relying on nonbanks should expect more questions on inclusion performance by geographic areas.

    • Closing a gap so large nonbank lenders meet community needs the same as how banks do

    • CRA style ratings for everyone and the performance will affect future license and applications

    • If a sponsor bank has embedded mortgage volume in New York, make sure your nonbank partner is adhering to the new guidelines

  • Mid-January, AFC and ICBA jointly pushed the Fed, FDIC, and OCC to update confidential supervisory information rules to allow banks to share limited details with partners for quicker remediation. Expect tighter guardrails and transparency in partnerships.

    • Faster fixes for consumers and safer programs

    • Any modernization is likely to come with clearer boundaries that formalize data-sharing protocols and have a need for paper trails that examiners can test

    • Sponsor banks should map which embedded partners truly need visibility to handle fixes and how to include it in your third party risk management. Map it early and it will make things smoother when responses come back or examiners ask questions.

  • OCC proposed January 8 to clarify national trust banks can handle certain non-fiduciary activities alongside fiduciary ones. Doesn't change chartering scope, but it matters for embedded custody, wallets, or trust-based rails.

    • OCC is cleaning up confusing language for trust banks

    • Supervisors have guidance on checking how each embedded use of a trust charter applies to “trust company and related activities”

    • If an embedded model leans on a trust bank for custody or wallets, document which flows are fiduciary vs which are not and why each fits OCC expectations before the next exam or comment period

  • RevitPay projected January 15 that embedded transaction volume could exceed $7 trillion annually by 2030, revenues tripling from recent levels. Big numbers on paper. But let's be honest: this growth only holds if sponsors prove strong governance, oversight, and loss models survive exams and rate shifts.​

    • The forecast reinforces that embedded is becoming the default way money moves inside software and platforms

    • Raises the bar for “acceptable” risk management, prompting regulators and boards to ask whether embedded programs are being run with the same control rigor

    • Ensure growth cases match what a conservative examiner and risk committee will actually sign off on

Embedded Finance is now being treated like real banking, not side revenue. Policy changes and rule clarifications/tweaks are all pushing sponsors to show they actually own pricing, partners, and risk. The teams that write that story down now, with clear files and clear accountability, will be the ones still growing when the next round of headlines hits.

Takeaway: Early 2026 Is About Proving Embedded Is Bank-Grade

Early 2026 is a reminder: treat embedded like core banking, not side revenue. Prove mortgage and credit products handle policy-driven rate shifts without breaking economics. Prove nonbank/platform partners meet CRA-style and community expectations as New York (and likely others) push nonbanks toward bank-like rules. Prove bank-fintech relationships have clear risk owners, solid controls, and enough info flow to fix issues quick instead of waiting for exams. Lead with compliance, build proof points early, and you'll move through the year smoother. The teams that do will keep growing embedded without unwanted attention.

From The Source

For those of you wanting a more in-depth look at the articles (and the links to them…)

On January 8, President Trump said that he was directing reps to buy $200 billion in mortgage bonds through Fannie and Freddie to pull rates and payments down; FHFA confirmed they'd execute

Rates did drop quick: average 30-year fixed to 5.99%, lowest in nearly three years. That gives embedded mortgage and housing-adjacent products a short affordability bump and more volume funneling in. Then January 9-12, he floated a one-year 10% cap on card APRs starting January 20 via Truth Social and comments. This proposal would need Congress to amend Truth in Lending, and banks are already warning it squeezes credit supply and economics if it becomes law. For BaaS or embedded built on cards, BNPL, or revolving credit, this combo reminds you: policy risk is baked into pricing now. Sponsors can't ignore the variability.

New York DFS finalized January 7-8: CRA-style community standards now apply to non-depository mortgage bankers originating 200+ home loans in the state the prior year, effective July 7, 2026. Covers fair access, geographic distribution, and serving low/moderate-income areas; basically brings big nonbanks closer to bank CRA. For embedded housing platforms or fintechs routing NY volume through nonbanks, distribution, product design, and marketing get looked at through a community lens, not just growth. Multi-state setups heavy on nonbanks? Start mapping where similar rules could pop up next.

Mid-January, AFC and ICBA sent a joint letter to the Fed, FDIC, and OCC urging them to update confidential supervisory information rules. Goal: let banks share limited, purpose-specific details with fintech/third-party partners who handle remediation or key operations. The argument is that old CSI rules slow fixes and weaken risk management when partners can't see enough to act. For BaaS and embedded, this signals regulators might allow more structured sharing with clear docs, controls, and guardrails. Expect tighter transparency standards in partnerships.

OCC issued a notice January 8 proposing to clarify that national trust banks can engage in certain non-fiduciary activities (tied to their business) alongside fiduciary ones. It's meant to align with the statute and eliminate confusion from older rules: no expansion or contraction of authority. Matters for embedded finance using trust structures for custody, wallets, or asset handling. If that's your setup, check how planned non-fiduciary pieces fit OCC expectations, and be ready to comment if the final rule shapes your rails.

For those structures, this is a signal to check how planned non-fiduciary services line up with OCC expectations and to be ready to comment if the final rule will shape how embedded rails are structured.

On January 15, RevitPay released a report projecting embedded finance transaction volume over $7 trillion annually by 2030, with revenues tripling from recent levels. They frame it as a growing pool reliant on banks/platforms for the rails. But let's be honest: commentary from banks and industry keeps saying this growth only sticks if sponsors show strong governance, clear third-party oversight, and loss models that hold across rate cycles and exams. Otherwise, the numbers stay on paper.

If you care about modern banking, detailed breakdowns of how financial institutions work with fintechs, and partnerships that actually perform, and you want access to candid conversations that usually stay inside the vault, this series is built for you.

Subscribe now to get the first episodes as soon as they drop and stay ahead of the next wave of bank-fintech moves.

Listen on your favorite podcast platform: listen.frominsidethevault.com

Watch full conversations and clips: watch.frominsidethevault.com

Get email recaps and future drops: subscribe.frominsidethevault.com

Keep Reading

No posts found