The Global Risks Alliance (GRA)
Payments are moving from batch and card rails to instant, always-on and increasingly cross-border services, built by banks, payment institutions, technology firms and public operators side by side. Speed and reach open new paths for failure: irrevocable settlement leaves less time to catch fraud, and more services depend on fewer processors and clouds. This page sets out how new payment services change operational, liquidity and fraud risk, how that risk is measured, and who decides what resilience is required.
GRA · Established · service level R0
Instant payments risk is the set of failure modes that open as payments move from batch and card rails to instant, always-on and cross-border services. Speed removes the buffer: fraud becomes irrevocable faster, liquidity is needed outside business hours and shared processors concentrate failure.
How a failure at one financial centre can spread through settlement links. Design links, not real dependencies.
New payment services and resilience. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Instant payments settle in seconds, round the clock and irrevocably. Batch cycles once gave banks hours to spot errors and fraud; that window now closes before any review, so controls must run inside the payment flow itself.
Round-the-clock settlement needs central bank money or prefunded positions at night and at weekends. Participants that size liquidity for business hours can be left short when a large outflow arrives while money markets are closed.
Many banks and payment firms rely on the same processors, card networks, messaging providers and clouds. An outage at one node stops payments at every institution behind it, and customers experience it as a failure of the brand they know.
Authorised push payment scams, mule accounts and account takeover migrate to whichever rail settles fastest with the weakest checks. Reimbursement rules and confirmation-of-payee services shift where losses fall, and fraud moves again.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| Authorised push payment fraud | Fraud in which a victim is tricked into authorising a payment to a criminal's account. |
| ISO 20022 | An international messaging standard for financial data that carries richer, structured payment information. |
| Systemically important payment system | A payment system whose failure could trigger or transmit disruption across the financial system. |
Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.
| Indicator | Unit | Cadence | Source type | Label |
|---|---|---|---|---|
| Payment system availability | percent of scheduled operating time | monthly | Operator availability logs | OFFICIAL |
| Time to recover a critical payment service | minutes from disruption to restoration | per incident | Operator and supervisor incident reports | REPORTED |
| Instant payment share of credit transfers | percent of transaction volume | quarterly | Central bank payment statistics | OFFICIAL |
| Authorised push payment fraud | losses in local currency; cases per million transactions | half-yearly | Industry fraud reporting | REPORTED |
| Intraday liquidity usage | peak share of available liquidity | daily | Settlement system records | OBSERVED |
| Payment rejection and settlement failure rate | percent of submitted payments | monthly | Operator records | OBSERVED |
| Concentration of payment processing | share of volume through the largest processors | annual | Overseer surveys | ESTIMATED |
The topic as a computational graph: named sources feed the methods, an evidence review gates the outputs, and the decisions stay with the authorities that hold them.
Fig. 02Nodes carry their type, two metric rows and a state; connectors show what feeds what; dashed frames name the owners; the timeline is the cadence. Design of the pipeline, not a record of a run.
Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.
| Source | What it gives | Licence class |
|---|---|---|
| CPMI Red Book statistics (BIS) | Payment, clearing and settlement volumes and values by country and by system | Free, attribution under BIS terms of use |
| ECB payment statistics | Euro area payment transactions by instrument and channel, including card and credit transfer fraud | Free, attribution under ECB terms |
| World Bank Global Payment Systems Survey | National payment infrastructure, regulation and oversight arrangements | Free and open, attribution |
| Instant payment operator statistics (Pix, UPI, FedNow, Faster Payments) | Volumes, values, availability and participation published by each operator | Public records, terms per operator |
| Industry fraud reports (UK Finance and peers) | Authorised and unauthorised payment fraud losses by type and channel | Free reports, terms per publisher |
| FSB and CPMI monitoring of the G20 cross-border payments roadmap | Progress against targets on cost, speed, access and transparency | Free, attribution under FSB and BIS terms |
| Central bank oversight reports | Assessments of systemically important payment systems against the PFMI | Public records, terms per central bank |
Fintech Nexus prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| Who decides | The question | Evidence this page serves |
|---|---|---|
| Central banks as operators and overseers | Whether an instant or cross-border system meets the PFMI, and on what terms new participants gain access | Availability and recovery records, liquidity usage, participant concentration |
| Payment system operators and scheme boards | What availability, recovery time and fraud controls to require of participants | Incident histories, fraud typologies, dependency maps of shared processors |
| Bank and payment institution boards | Whether liquidity, fraud controls and third-party contracts are sized for round-the-clock settlement | Intraday liquidity scenarios, fraud loss trends by channel, recovery-time evidence |
| Conduct and consumer protection authorities | How losses from authorised push payment fraud are shared between sending and receiving firms | Fraud reporting by firm and type, effects of confirmation-of-payee and reimbursement rules |
Payment systems combine front-end services that customers use, such as cards, apps and wallets, with clearing and settlement arrangements that move money between institutions, ultimately settled in central bank money for systemically important systems. Innovation has concentrated at the front end, while settlement has moved towards real-time gross settlement and instant payment schemes that operate continuously.
Failures arise from operational outages, cyberattacks, liquidity shortfalls at participants and fraud. Instant, irrevocable payments make authorized push payment fraud more attractive, because funds can be moved on before anyone notices. Concentration in a few processors, networks or wallets means that one outage can stop commerce across a region. Non-bank providers hold customer funds under safeguarding rules that vary by jurisdiction.
Cross-border payments involve chains of correspondent banks, which add cost and delay; new arrangements that link instant payment systems or use new settlement technology promise improvements, but raise questions about governance, legal certainty and compliance with financial crime rules.
Offline resilience is an emerging question: as cash use declines in some countries, authorities and researchers are asking how payments will work during prolonged power or network outages, and which fallback arrangements are realistic.
Volumes, values and outage records; granular data are held by operators and central banks.
Simulating how participant failures or outages spread through payment networks.
Studying past outages and fraud waves; public incident reports are often limited.
Limits: much payment data is confidential, so research relies on aggregates and published incidents, and simulations depend on assumptions about participant behaviour.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.