The $1.4 Trillion Settlement Gap Is an Operating-Model Problem
The latest global evidence shows that most foreign-exchange settlement is protected or mitigated, yet a material principal-risk tail remains. Banks now need a trade-level control model that makes the safest practicable settlement route the default and treats every exception as a governed exposure.
The Bank for International Settlements’ June 2026 analysis gives the industry its clearest recent view of the settlement problem. In April 2025, just over $5.2 trillion of average daily FX settlement used payment versus payment (PvP), which eliminates settlement risk. A further $7.6 trillion used intragroup settlement, pre-settlement netting or bank-account arrangements with timing controls. The remaining $1.4 trillion, equal to 10% of daily settlement measured by the survey, was settled gross bilaterally and therefore exposed to full principal risk.
That headline should not be read as evidence that controls have failed. The BIS found that 90% of average daily settlement used methods that eliminate or mitigate risk, and the gross-bilateral share has fallen materially compared with a methodologically similar 2006 survey. The management question is narrower and more actionable: why does the residual exposure remain, where is it concentrated, and how can a bank reduce it without creating an unmanageable liquidity or operational burden?
Read the survey as a control map, not a market statistic
Settlement risk is unusual because the value at risk is the principal, not merely the replacement cost of a failed trade. One currency can be paid away with finality while the counter-currency is not received. The exposure may last only hours, but its size can be large and its transmission can be rapid through nostro accounts, correspondent relationships and liquidity commitments.
The BIS hierarchy is therefore useful as an operating taxonomy. PvP sits at the top because it makes final transfer of one currency conditional on final transfer of the other. Pre-settlement netting can sharply reduce the amount that must be paid, while timing controls and intragroup arrangements mitigate particular pathways. Gross bilateral settlement leaves the full exchange of principal exposed. These methods are not interchangeable labels; they represent different residual risks that should attract different approval, limit and evidence requirements.
The survey also identifies the main practical causes of gross settlement: a counterparty may not have access to PvP, or the currency pair or trade type may be ineligible. Those are portfolio attributes. A bank can map them before value date, price their operational consequences, and direct sales, trading and onboarding teams toward safer routes rather than discovering the limitation during settlement.
Build an eligibility and exposure view at trade level
A credible control model begins with a common settlement data set. Each trade should carry the legal entities, currencies, product type, expected value date, settlement route, provider, cutoff times, nostro accounts and netting set. It should also record why a higher-ranked method was not used. Free-text explanations are difficult to govern, so exception reasons should come from a controlled taxonomy such as currency ineligible, product ineligible, counterparty access unavailable, cutoff missed, documentation incomplete or temporary service disruption.
The next layer is an exposure clock. Risk teams need to know when the sold currency becomes irrevocable, when the bought currency is expected, and when an exception becomes overdue. Aggregation should work by counterparty, group, currency, nostro, country of settlement and business line. A static end-of-day report is insufficient for an exposure that changes through payment windows; intraday monitoring and escalation are part of the control, not merely an operational dashboard.
Limits should distinguish eliminated, mitigated and unmitigated settlement. A gross bilateral obligation should consume a principal-risk limit for the full amount during the exposure window. A netted obligation should be measured after legally enforceable netting, with the underlying calculation and cutoffs evidenced. Intragroup flows still require explicit assumptions about entity separateness, funding and resolution constraints. This produces a more faithful risk view than applying one broad settlement factor to all FX activity.
Make the risk waterfall executable
The Global Foreign Exchange Committee’s 2025 update strengthened five of the FX Global Code’s 55 principles, including its treatment of settlement risk. The practical direction is a waterfall: eliminate risk through PvP where practicable, reduce obligations through netting when PvP is unavailable, apply other risk-reducing controls, and minimise gross bilateral settlement. A bank should translate that hierarchy into routing logic and exception authorities rather than leave it as policy prose.
1. Default eligible flow to PvP
The execution and confirmation stack should identify PvP eligibility as early as possible and select it automatically when the entities, currencies, product and value date qualify. Onboarding must include the relevant settlement-member or third-party access, static data and testing. A manual choice to bypass PvP should create an approval record and a measurable exception.
2. Industrialise netting for the remainder
Netting is valuable only when legal enforceability, data quality and operational deadlines are reliable. The BIS found that pre-settlement netting reduced about $2.2 trillion of average daily obligations to $337 billion. That illustrates the scale of possible compression, but a bank still needs matched trade populations, agreed calculation rules, dispute workflows, cutoff discipline and a controlled path from the net result to actual payment.
3. Constrain gross bilateral settlement
Some flow will remain outside PvP and netting. The objective is not to relabel that exposure as unavoidable; it is to make it visible, limited and time-bound. Controls can include counterparty limits, prefunding rules, sequencing of payments, tighter release authorities, daylight-overdraft constraints, confirmation deadlines and heightened monitoring for late receipt. Repeated exceptions should trigger an onboarding, product-design or commercial decision.
Separate settlement safety from liquidity efficiency
CLSSettlement reports that it settles more than $8 trillion of payments a day in 18 currencies and that multilateral netting reduces funding requirements by more than 96%. Those figures show why risk reduction and liquidity efficiency can reinforce each other. They should not, however, lead banks to assume that every flow is eligible or that adding PvP access is operationally trivial.
The CPMI’s work on wider PvP adoption describes structural barriers including access to payment systems and central-bank accounts, differences in operating hours and limited overlap between real-time gross-settlement systems. This means the target architecture must include local-currency funding, cutoff calendars, contingency liquidity and service-provider dependencies. A route that is theoretically safer can fail in practice if the bank cannot fund it predictably.
Treasury and operations should therefore model the peak funding requirement, timing of pay-ins and pay-outs, intraday buffers and fallback capacity for each settlement route. Stress scenarios should combine a provider outage, reduced market liquidity, a late counterparty and a high-volume value date. The decision metric is not simply gross settlement value; it is the amount and duration of unsecured principal exposure after risk-reducing controls, considered alongside the liquidity needed to use those controls.
Govern counterparties and providers as part of the trade
Settlement choice is often shaped by counterparties, correspondents and infrastructure providers. Due diligence should cover service eligibility, operating windows, matching rules, resilience, recovery objectives, data access, incident notification and exit support. Contracts need to preserve the bank’s ability to obtain transaction-level evidence and to change routes without losing the legal or operational basis for netting.
Management information should connect settlement outcomes to commercial ownership. Useful measures include the share and value settled through PvP, the amount compressed by netting, gross bilateral exposure by reason, average exposure duration, missed cutoffs, late receipts, failed settlements and repeated counterparty exceptions. Business heads should see both the risk and the revenue associated with exceptions so that remediation choices are economically informed.
Treat data lineage and control testing as first-class evidence
A settlement-risk dashboard is only as reliable as the joins between execution, confirmation, payment and account data. Control owners should document the source and transformation of each critical field, including value date, finality timestamp, netting status and receipt time. Reconciliations must identify omitted trades, duplicate instructions, late amendments and payments that were rerouted after the original confirmation. Sampling a clean population will understate the very exceptions the framework is meant to expose.
Testing should combine design and outcome evidence. A design test asks whether eligible trades are automatically directed to the intended route and whether bypass requires the correct approval. An outcome test asks whether payments actually settled through that route, whether the counter-leg arrived within the expected window and whether the exposure was measured correctly. Persistent differences between designed and realised routes should be treated as control failures, not reporting noise.
Client experience matters as well. A bank may be able to reduce risk by changing documentation, confirmation timing, cutoff communication or access arrangements for customers that repeatedly settle gross. Segmenting clients by exposure and remediation feasibility lets relationship teams focus on high-value improvements. The message is operational rather than punitive: a safer route can reduce payment uncertainty and manual investigation for both parties.
A 90-day implementation sequence
In the first 30 days, establish the taxonomy and baseline. Reconcile trade, confirmation, payment and nostro data for a representative period; classify every route; and identify the largest gross-bilateral concentrations. Validate the legal basis for netting and the integrity of settlement timestamps before relying on aggregated results.
In days 31 to 60, redesign defaults and authorities. Configure eligibility checks, route eligible flow to PvP, create an exception-reason hierarchy, set principal-risk limits and define escalation times. Prioritise counterparties and currencies where a small number of onboarding or documentation changes can remove a large exposure.
In days 61 to 90, rehearse and govern. Run high-volume and provider-outage scenarios, prove that treasury can meet funding windows, and test how operations detect and escalate late receipt. Present the first settlement-risk scorecard to the relevant trading, treasury, operations and risk committees. The outcome should be a controlled book of residual exposures, not a one-off remediation list.
The strategic conclusion
The latest evidence reframes FX settlement risk as a portfolio-management discipline. The market has made substantial progress, but the remaining exposure is too large to treat as an operational footnote. Banks that make the settlement hierarchy executable can reduce principal risk, improve liquidity visibility and turn adherence to market standards into demonstrable day-to-day control.
Frequently asked questions
What is FX settlement risk?
It is the risk that a firm pays the currency it sold but does not receive the currency it bought. Because the principal amount can be lost, the exposure is different from ordinary replacement-cost risk.
How does payment versus payment reduce the risk?
PvP makes final transfer of one currency conditional on final transfer of the other. When both legs settle together, the principal-risk gap between payment and receipt is eliminated.
Does netting eliminate FX settlement risk?
No. Pre-settlement netting reduces the amount that must be paid, which can greatly reduce exposure, but the remaining net payment still needs a risk-controlled settlement method.
Why do firms still use gross bilateral settlement?
The BIS identified lack of counterparty access and ineligible currency pairs or trade types as major reasons. Missed cutoffs, incomplete documentation and temporary service constraints can also create exceptions.
What should a bank measure first?
Start with the value and duration of gross bilateral exposure by counterparty, currency, product and exception reason, then add PvP usage, netting compression, missed cutoffs and settlement-fail indicators.
References
1: BIS Quarterly Review: Uncovering FX settlement risk, June 2026
2: Global Foreign Exchange Committee: 2025 FX Global Code update
3: CPMI: Facilitating increased adoption of payment versus payment