Why U.S. payments are slow

In May 2008, the U.K.'s Faster Payments Service went live, providing near-instant settlement for retail payments among the participating U.K. banks. In November 2017, the EU launched SEPA Instant, providing the same capability across the eurozone. In November of that year, The Clearing House launched RTP in the United States. In July 2023, the Federal Reserve launched FedNow. By the standard arithmetic, the United States got real-time consumer payments roughly fifteen years after the U.K. and roughly six years after the eurozone. The lag is unusual for a country that has been on the leading edge of financial-technology development in most other respects, and it produces a question worth asking: what specifically was the U.S. doing differently?

This essay argues that the U.S. lag was institutional rather than technological, that the explanations have to do with the structure of U.S. banking and the absence of the regulatory mandates that drove faster adoption elsewhere, and that the arrival of FedNow has accelerated the timeline modestly but has not produced the full convergence with U.K./EU practice that some commentators expected. The conclusion is that the U.S. payment system continues to be shaped by the same institutional dynamics that produced the original lag, and that the next decade of evolution will look different from the U.K. and EU trajectories.

The institutional explanation

The U.K. moved early on real-time payments because the Office of Fair Trading, then the Competition and Markets Authority, mandated it under a 2005 review of U.K. payment systems. The mandate was binding on the U.K.'s clearing banks and produced Faster Payments as a regulatory-driven utility rather than as a competitive-market product. The eurozone's SEPA Instant emerged from a parallel regulatory process at the EU level, coordinated through the European Payments Council and the European Central Bank. In both cases, the central question — whether to build a real-time consumer-payment rail — was answered by a regulator rather than by market participants.

The United States lacks a comparable regulatory mandate authority. The Federal Reserve operates Fedwire and FedACH; it has no statutory power to require U.S. banks to adopt a particular consumer-payment service. The Consumer Financial Protection Bureau has authority over consumer-protection rules but not over the construction of payment-system infrastructure. The state banking regulators have parallel authority but not industry-wide coordinating power. The result is that the construction of a real-time consumer-payment rail in the U.S. has been the result of voluntary industry coordination (RTP, launched in 2017 by the bank-owned Clearing House) and voluntary Fed initiative (FedNow, launched in 2023 after years of deliberation), without the binding force that drove the U.K. and EU implementations.

The voluntary character of the U.S. effort had several consequences. Adoption was slower because individual banks could decline to participate, particularly smaller banks for whom the cost of connection exceeded the early customer demand. The RTP network reached most large U.S. bank accounts within several years of launch but was not universal; FedNow's launch was framed explicitly as ensuring access for smaller institutions. The bilateral negotiation among large-bank participants over RTP's governance and pricing slowed deployment. The Fed's deliberation over whether to build FedNow at all — initiated in 2013 with the Faster Payments Task Force — took the better part of a decade to conclude.

The structural reasons for slow industry coordination

Three structural features of U.S. banking made voluntary coordination on real-time payments slower than its U.K. and EU equivalents.

First, the U.S. system has roughly 9,000 depository institutions, against fewer than 400 in the U.K. and several thousand across the eurozone with substantial concentration in the largest. Coordinating a new payment rail across 9,000 institutions, with varying technology capabilities and varying business interests, is genuinely harder than coordinating across a smaller and more concentrated set. The U.S. could not adopt the U.K. approach of "mandate the clearing banks and the rest will follow" because there is no equivalent small group of clearing banks; the deposit system is meaningfully distributed.

Second, the existing card networks — Visa, Mastercard, American Express, Discover — have served much of the use case that real-time consumer payments address in other markets. In the U.K. and EU, card payments are routine for in-store purchases but less dominant for account-to-account transfers; the gap that Faster Payments and SEPA Instant filled was meaningful. In the U.S., debit cards have been heavily used for both purchase and (through Visa Direct and Mastercard Send) for instant transfers, blunting the market demand for a separate real-time-payments rail. The CFPB's research on consumer payment behavior consistently shows U.S. consumers using cards for transactions that European consumers would handle through bank transfers.

Third, the U.S. ACH network had been gradually accelerating throughout this period. Same-day ACH was launched in three phases between 2016 and 2018, with additional same-day windows added in 2021 and the per-transaction limit raised multiple times. While same-day ACH is not real-time in the technical sense — it operates in batch settlement windows rather than instant continuous settlement — the practical experience for many consumer use cases (payroll, bill pay, P2P transfers through Zelle that use same-day ACH or RTP as the back-end rail) was that ACH had become "fast enough" for the most common needs. The marginal benefit of moving from same-day ACH to real-time-final settlement was smaller in the U.S. context than in countries where ACH had not been accelerated comparably.

What FedNow has and has not changed

FedNow's July 2023 launch brought real-time payments to a substantially broader base of U.S. depository institutions than RTP had reached. The Fed's connection model — through the master accounts that all depository institutions hold at their Federal Reserve Banks — meant that any insured depository institution could in principle connect to FedNow with relatively modest operational changes. Adoption has grown steadily: the Federal Reserve reported over 1,400 participating institutions by 2025, with the number continuing to grow.

What FedNow has not done is produce the discontinuous shift in consumer payment behavior that some commentators predicted. Consumer use of real-time payments — through RTP, FedNow, or the bank apps that route through them — has grown but remains substantially below the level of card-network or ACH use for comparable purposes. The reasons for the slower-than-expected adoption are multiple:

• Most consumer-facing apps (Zelle in particular) have offered substantially real-time experience for years through a combination of same-day ACH and RTP, blunting the marginal benefit of a pure-RTP or FedNow back-end.
• The consumer-protection framework for real-time payments is narrower than for ACH and cards — particularly for authorized-but-fraudulent payments, see Zelle, Venmo, Cash App — making banks reluctant to push customers toward real-time rails for higher-stakes use cases.
• Banks have priced real-time payments above ACH on a per-transaction basis, providing incentive to continue routing through ACH where the speed difference is not consequential.
• Business-to-business adoption, where the real-time-payments value proposition is theoretically strongest, has been slow because business workflows and accounting systems are built around batch processing.

The U.K. precedent is instructive in both directions

Several features of the U.K. Faster Payments experience are worth noting because they predict what the U.S. trajectory may look like.

First, U.K. consumer adoption of Faster Payments did grow steadily but more gradually than initial projections suggested; the rail required several years to become routine and longer to displace alternative payment methods for some use cases. The rapid early growth was concentrated in account-to-account transfers between consumers' own accounts; the broader migration of bill-pay and merchant-payment use cases took most of a decade.

Second, U.K. authorized-push-payment fraud became a substantial consumer-protection problem as Faster Payments adoption grew, prompting the 2019 Contingent Reimbursement Model and the 2024 mandatory reimbursement regime under the Payment Systems Regulator. The U.S. has not adopted an equivalent framework; the fraud-protection question that the U.K. has now addressed in two iterations is, in the U.S., still substantially unresolved.

Third, the broader U.K. open-banking framework — built on top of Faster Payments as a rail — produced the third-party-payment-initiation services and the account-aggregation ecosystem that European fintech now relies on. The U.S. is replicating this structure through the CFPB's Section 1033 rule (see open banking in the U.S.), with implementation phased through 2030. The U.K. timeline suggests that the full effects of this kind of framework take years to play out.

The counter-argument

The strongest counter-argument to the institutional thesis is that the U.S. simply did not need real-time payments as urgently as the U.K. and EU did, and that the slower trajectory reflects a rational allocation of market resources rather than a regulatory failure. In this view, the U.S. card and same-day-ACH infrastructure was substantially better than the U.K. and EU equivalents at the time those countries adopted real-time rails; the marginal benefit of adding instant settlement on top of an already-fast system was smaller; and the deliberate U.S. approach has produced a more competitive payments market with multiple rails (RTP, FedNow, Zelle, card networks) operating in parallel.

There is force to this. The U.S. payments system, viewed holistically, offers consumers more channels and more competitive pricing than the U.K. or EU equivalents at the time real-time rails were introduced abroad. The fact that the U.S. arrived at real-time payments later does not by itself mean the U.S. is behind in consumer payment-system functionality; it means the trajectory was different.

The counter-argument starts to weaken when one looks at the specific use cases where the U.S. has lagged most visibly. International remittances — a major use case for which competitive U.S. providers exist but not on bank-direct rails — remain expensive and slow. Account-to-account bill payment continues to depend on ACH or on Zelle-front-ended bank rails, with substantial operational complexity. Business-to-business payments at scale are still dominated by ACH and (for large-value) wires, with real-time alternatives only beginning to mature. The U.S. payments system is sophisticated, but the specific gaps it has are gaps that the U.K. and EU started closing earlier.

What the next decade looks like

If the institutional explanation is right, the next decade of U.S. payment-system evolution will look more like the past decade than like a discontinuous shift. Real-time payments will grow steadily through both RTP and FedNow, with the per-transaction limit rising and the use case coverage expanding. The CFPB's Section 1033 rule will produce a more European-style open-banking ecosystem with phased compliance through 2030. Consumer-protection frameworks for real-time payments will likely catch up to U.K. practice, either through CFPB rulemaking or through litigation-driven bank-side voluntary changes. The card networks will remain dominant for in-person and many online purchases but face increasing competition from pay-by-bank arrangements built on real-time rails. The institutional dynamics that produced the original lag — multiple-charter structure, voluntary industry coordination, limited mandate authority — will continue to slow the U.S. trajectory relative to the U.K. and EU equivalents, but the gap should narrow over time.

The deeper point is that "U.S. payments are slow" is a comparison to a specific benchmark — the binding-mandate-driven adoption of real-time rails in markets with smaller, more concentrated bank systems and stronger central regulatory authority. The U.S. payment system is not slow in absolute terms; it operates at scale with substantial sophistication. What it lacks is the binding-coordination mechanism that drove faster adoption elsewhere, and the institutional structure that would have permitted such coordination. The trajectory is moving, but the structural reasons it moved slowly remain in place.