The US has more payment rails in production use than any comparable economy in the world, four widely used systems for domestic funds transfer, two live card networks that account for the majority of consumer spending, and a cross-border correspondent-banking layer that still moves most of the country’s international payments. That complexity is the backdrop for every US fintech’s infrastructure decision. The global payments market was valued at roughly $2.4 trillion in 2024 and is projected to exceed $4 trillion by 2030, according to Grand View Research, and the US share of that spend continues to rise.
The four US domestic rails
The US has four distinct rails for moving money between domestic bank accounts: ACH, wire (Fedwire and CHIPS), the Federal Reserve’s FedNow service, and The Clearing House’s RTP network. Each has different cost, speed, and operational characteristics, and each is used for different categories of transactions inside a typical US financial product.
| Rail | Typical settlement speed | Typical cost | Dominant use cases |
|---|---|---|---|
| ACH | Same-day to next-day | Low (cents) | Payroll, bills, recurring payments |
| Wire (Fedwire/CHIPS) | Minutes | Moderate ($15-$50) | High-value, time-sensitive, M&A, real estate |
| FedNow | Instant (24/7) | Low | Consumer payments, emergency disbursements |
| RTP | Instant (24/7) | Low | Consumer and B2B instant payments |
Source: Grand View Research; see the Grand View digital payments report.
No single rail is the right answer for every use case. The fintech design decision is which combination to use for which category of transaction, and that decision has become more complex since the introduction of FedNow in 2023.
Why the US rail count is different from other countries
Most developed economies have converged on one or two payment rails: typically an instant rail operated by the central bank and a card network. The US has kept multiple rails in parallel for two structural reasons. The first is that the card networks are so large and so profitable that neither the banks that issue their cards nor the networks themselves have had strong commercial incentives to compress the market. The second is that the Federal Reserve and The Clearing House operate independently, which has produced two parallel real-time systems rather than one consolidated one.
The practical consequence is that US fintechs have to make infrastructure choices that their European or Australian counterparts rarely have to make. A UK fintech can default to Faster Payments; a US fintech must choose between ACH, wires, FedNow, RTP, card rails, and sometimes a proprietary bank rail, for each transaction type.
The card network layer
Card networks, Visa and Mastercard, with American Express and Discover as smaller but meaningful participants, handle the majority of US consumer spending and are the backbone of most US fintech consumer revenue models. The card network economics are unusual: merchants pay interchange and assessment fees that are meaningfully higher than direct bank-to-bank transfer rails, and most of the interchange is distributed back to the card issuer, typically a bank or a fintech sponsor bank, in the form of revenue.
For consumer fintechs, card interchange remains the single largest revenue line, even as it has declined as a share of total revenue. The broader dynamic of how these revenue lines have rebalanced was covered in our piece on the role of venture capital in fintech growth.
Cross-border payments infrastructure
The cross-border payments layer for the US is still largely the correspondent-banking network, supplemented by a set of fintech intermediaries that aggregate and route payments more efficiently. The Federal Reserve’s international wire transfer data shows that wire volumes and values continue to grow, but the fintech-intermediated share of cross-border consumer and SMB payments has grown faster.
The direction of innovation in US cross-border payments is split. Stablecoins and tokenized deposits are capturing a small but growing share of cross-border B2B payments, particularly between US dollar holders. At the same time, the Society for Worldwide Interbank Financial Telecommunication (SWIFT) has continued to upgrade its messaging and settlement services, shortening cross-border settlement times without fundamentally changing the correspondent-banking model.
How fintechs combine rails in production
A typical US fintech product combines three or four rails in a single user experience. Consumer funding and withdrawals usually happen over ACH. Instant payouts, for earned-wage access, insurance claims, or marketplace earnings, typically run over RTP or FedNow. High-value SMB and B2B payments use wires or, increasingly, same-day ACH. Consumer spending and merchant acceptance run over the card networks.
The operational complexity of combining these rails is the reason so many US fintechs partner with a single payments-infrastructure provider, companies like Stripe, Adyen, Marqeta, Alloy, Plaid, Modern Treasury, Finix, Dwolla, that handle the integration layer across rails. That vendor layer is the part of the US fintech ecosystem that has most consistently attracted venture capital, and it is where the most visible commercial battles have played out over the last five years. The broader digital-banking context for these shifts is covered in our reporting on why digital banking adoption is accelerating among SMEs.
What regulation has done to the US stack
US regulation has shaped payments infrastructure more than any market force. The Bank Secrecy Act, the Electronic Fund Transfer Act, Regulation II on debit interchange, the CFPB’s rules on money transmission, and state money-transmitter laws together define what a US fintech can and cannot do with each rail. The compliance burden is material and is one reason most fintechs route their payments through regulated partners rather than holding full licences themselves.
The regulatory trend in 2024 and 2025 has been toward tighter fraud-liability rules across real-time rails, expanded AML obligations for non-bank payments companies, and continued pressure on debit-card interchange. The effect has been to raise the bar for running payments infrastructure in-house, which has benefited the large infrastructure vendors at the expense of fintechs that had been trying to hold more of the stack internally. The strategic shift is part of the priority realignment we covered in our piece on why fintech is becoming a strategic priority for financial institutions.
What this means for US fintech operators
For US fintech operators, the infrastructure decision is now the single most consequential early-stage product decision. A fintech that picks the wrong combination of rails or the wrong infrastructure partner will find itself either over-paying for transaction processing or unable to offer competitive features, instant payouts, for example, or low-friction cross-border transfers.
The best 2025 operators treat payments infrastructure as a first-order strategic decision rather than a procurement task. That usually means deciding early whether to hold any licences directly, which infrastructure vendor to standardise on, and which rails to offer customers by default. The competitive dynamics around those choices sit inside the broader pattern we analysed in our piece on how fintech is reshaping competition in financial services.
The longer arc
US payments infrastructure is not converging; it is diversifying. Instead of collapsing to one or two rails, the US market is adding new rails and new infrastructure layers on top of the existing ones. That is expensive for fintechs to navigate but also creates competitive opportunity, because the winners will be the operators who can stitch together the rails into a product experience that competitors cannot replicate easily.
