Four developments this week touch the same nerve: how fintech products gain access to the bank partners, payment rails, and scaled financial products they depend on, and who sets the terms. The FDIC is drafting a body that would certify vendors once and let many banks reuse the result. Swift went live at Bank of America and JPMorgan with a cross-border framework that fixes fees upfront. RTP and FedNow loosened their own rules to reach across borders through intermediaries. And Goldman paid up to $2.25 billion for NEOS to buy a scaled ETF franchise rather than build one. Each layer is being certified, reopened, or bought, and the question is what you build yourself versus obtain through a standard, a network, an intermediary, or an acquisition.
FDIC drafts a shared certification body for bank technology vendors
The FDIC is in early talks with banking and fintech trade groups to create an independent nonprofit that would certify technology vendors against a published baseline of third-party risk standards. A vendor would be assessed once, keep the assessment current, and present the certificate to multiple banks instead of repeating a manual review with each one. The FDIC is expected to provide seed funding, participation would be voluntary, and each bank keeps its own obligation to monitor vendors. Collaborators include the ABA, ICBA, Bank Policy Institute, the Financial Technology Association, the American Fintech Council, and the existing Coalition for Financial Ecosystem Standards. The effort follows the Synapse collapse.
Source: https://www.americanbanker.com
Why it matters
Bank partnership diligence has been a private, per-bank process: sponsor banks often repeat substantial parts of the same vendor review, and the fintech re-answers overlapping questionnaires for each one. A shared certification turns that repeated cost into a portable credential and, in doing so, defines a public baseline for what “adequate controls” means. The result is a shared reference point for adequate controls alongside each bank’s own diligence. Certification does not replace onboarding: participation is voluntary and each bank keeps its monitoring duty, so the certificate is a reusable floor, not a pass to access.
The second effect is architectural. Once a baseline is published, control evidence has to be produced continuously, not assembled for each audit. Access permissions, reconciliation records, and monitoring logs become artifacts kept current for recertification, which favors products that already generate that evidence as a byproduct of how they run. Vendors without that instrumentation face rework before they can be certified once.
What teams should do
This applies to BaaS providers, banking and payments infrastructure vendors, and any fintech that sells into or partners with FDIC-supervised banks.
- Map your current diligence answers to the emerging standard’s control categories, so you can see the gap before certification exists rather than during a sales cycle.
- Instrument control evidence (access logs, reconciliation, monitoring) to refresh continuously, since a certificate that must stay current rewards always-on evidence over point-in-time reports.
- Assign an owner for recertification who tracks standard changes, the way you track SOC 2 renewal, before it becomes an unmanaged obligation.
- Preserve an audit trail that a third-party assessor can read without your engineers narrating it, because portable certification means outsiders review your records.
- Treat the certificate as a sales asset once available, and prepare community-bank materials that lead with it, since smaller banks are the intended beneficiaries.
Swift goes live at Bank of America and JPMorgan with a cross-border retail framework
Swift activated a new international retail transfer framework, with Bank of America and JPMorgan among the first US banks live on it. The framework shows the sender fees and exchange rates before a transfer is confirmed, guarantees the beneficiary receives the full amount without deductions, and settles within minutes where local rails support it. It adds a fixed-fee option as an alternative to variable correspondent-bank rates. About 60 banks across 25 countries back the initiative, with initial corridors including Australia, Brazil, China, India, South Africa, South Korea, Spain, and Turkey. Swift positions the framework as a way for banks to compete in small-value transfers they had largely ceded to non-bank players.
Source: https://www.americanbanker.com
Why it matters
Small-value cross-border transfers were ceded to non-bank players years ago, because correspondent banking made the margins too thin and the experience too opaque for banks to compete. What Swift is testing is a full customer proposition, not a faster pipe: fee shown before sending, exchange rate known in advance, and the beneficiary credited the full amount. Those are promises to the end user, and a bank can only make them if the fee and FX are fixed before the payment leaves, not discovered along the way.
The demanding part is standing behind that guarantee across banks the sender’s institution does not control. A promised amount has to survive every correspondent hop, which means the commitment is only as good as the weakest participant’s adherence to the framework. For teams, the lesson is that a cross-border product now competes on the outcome it can commit to before the payment is sent, and that commitment is an operational contract with the whole chain, not a feature of any one leg.
What teams should do
This applies to remittance apps, cross-border payment providers, and fintechs weighing how to deliver fast, transparent international transfers.
- Reassess your cross-border roadmap against a bank-network option that now offers upfront pricing and full-amount delivery, before committing to any single settlement architecture.
- Build fee and FX resolution to happen before a payment is sent, since guaranteeing the received amount means no deductions can appear mid-route.
- Compare coverage market by market against your current provider, since the framework’s guarantees only apply where it is live and local rails support them.
- Preserve structured payment data across every hop, because a guaranteed-amount promise breaks the moment fee or FX fields are stripped between systems.
- Instrument settlement time per corridor, since “within minutes” depends on local rails and your product should surface realistic delivery windows to users.
RTP and FedNow loosen rules to reach across borders
The Clearing House’s RTP Network is implementing a rule change later this year that would let one leg of a transaction involve a foreign bank. Separately, the Federal Reserve proposed amending Regulation J so FedNow participants can use intermediaries other than Reserve Banks, which would support the international leg of a cross-border payment while FedNow settles the domestic leg. FedNow would still settle only between eligible US participants. As of mid-August, industry groups including Wise, Visa, and Stripe have backed the FedNow proposal and urged the board to move quickly.
Source: https://www.americanbanker.com
Why it matters
US instant rails were built inward, for domestic transfers between US institutions. Opening them to a foreign leg through intermediaries does not make them cross-border rails; it makes them one component the fintech has to orchestrate alongside a separate international leg. Unlike a framework that promises the end user a fixed outcome, this gives teams a fast domestic settlement and leaves the rest of the chain to assemble.
That orchestration is where the product risk concentrates. The domestic leg can clear in seconds while funds sit in a post-arrival compliance hold, or while payment data is stripped crossing into another country’s format, and the two legs can settle on different timelines. The team owns stitching them into one coherent status for the customer, and owns the failure case where one leg completes and the other does not.
What teams should do
This applies to payroll and payout platforms, treasury products, and B2B fintechs moving money internationally over instant rails.
- Design reversal and exception handling for the case where the domestic leg settles instantly but the foreign leg fails, since irrevocable domestic settlement narrows your recovery options.
- Reconcile the two legs on separate timelines, so a completed domestic settlement is not shown to the customer as a finished payment while the foreign leg is still pending.
- Instrument each seam outside the rail (FX, post-arrival compliance hold, data truncation), since the instant domestic leg hides these delays from your dashboards.
- Evaluate intermediary banks for the foreign leg on settlement time and data fidelity, and treat that choice as a core dependency rather than a back-office detail.
- Compare the RTP foreign-leg model against the Regulation J intermediary model for your corridors, since the two expose different integration and settlement constraints.
Goldman Sachs buys NEOS for up to $2.25 billion to expand active ETFs
Goldman Sachs agreed to acquire NEOS Investments for as much as $2.25 billion in cash and equity. Founded in 2022 and based in Westport, Connecticut, NEOS manages about $30 billion across 19 options-based income ETFs, built on systematic covered-call and put-write strategies wrapped around major equity indices. The deal is expected to close in the first quarter of 2027 and would push Goldman Sachs Asset Management’s active ETF assets to roughly $130 billion, making it a top-eight active ETF manager. It follows Goldman’s roughly $2 billion purchase of Innovator Capital Management about a year earlier.
Source: https://www.reuters.com
Why it matters
An incumbent with its own asset-management platform paid a large sum for a four-year-old issuer rather than build the equivalent in-house. The strategies themselves, covered-call and put-write income wrappers, are well understood. What NEOS had built was a scaled franchise: $30 billion across 19 funds, established asset flows, a productized line, and the operating capability to run it. Goldman bought the assembled result, which is the part that takes years to compound even when the underlying strategy is replicable.
The read-across for wealthtech is where product value accumulates. A replicable strategy on its own offers limited defensibility; more value can accumulate in the franchise around it: fund operations, tax-lot handling, servicing, and established asset flows. Acquisition value depends on the quality and durability of those flows and that operating machinery, not on AUM alone.
What teams should do
This applies to wealthtech platforms, brokerages, robo-advisors, and firms building or distributing packaged investment products.
- Identify which part of your product is the strategy and which is the franchise around it, and invest in the layer an acquirer cannot easily replicate.
- Build fund and portfolio operations (tax-lot accounting, corporate actions, rebalancing) to institutional standards, since packaging is where option-income products succeed or fail operationally.
- Instrument distribution and flow economics per channel (cost to acquire assets, retention, flow stability), since acquisition value depends on the durability of flows, not AUM alone.
- Preserve the traceability of each product’s tax treatment and outcome logic, so the after-tax claims that attract investors can be substantiated under review.
- Assess build-versus-partner for complex strategies, since replicating a productized fund line in-house may cost more than integrating an existing one.
Closing insight
Run one review this week: for every bank partner, payment rail, intermediary, and external product your company depends on, identify who sets the access terms and what happens if those terms change. This week showed each of those being certified, reopened, or bought, sometimes with several approaches emerging at once, as seen around Swift, RTP, and FedNow.
Teams that can switch providers, prove their controls, and preserve payment data across integrations are less exposed when access terms move. In our work with fintech teams at Itexus, that portability usually comes down to unglamorous groundwork: control evidence generated as the system runs, structured data that survives every handoff, settlement instrumented per corridor, and operations that do not depend on a single provider staying still. The products that last keep the ledger, the evidence, and the customer relationship on their own side, whatever they rent or buy on the other.