ISO 20022 Statistics 2026
By Axis Intelligence Research
Co-author: Sarah Davis | Last updated: August 25, 2026 | License: CC BY 4.0
Swift observed 97% of cross-border payment instructions in ISO 20022 format on 24 November 2025, the day after the MT coexistence period closed. But only 13% of the world’s real-time gross settlement systems are actually aligned with the CPMI harmonisation requirements that make the data usable, against 79% that have migrated or plan to.
Quick Answer
The MT to ISO 20022 coexistence period for cross-border payments ran 32.1 months, from 20 March 2023 to 22 November 2025, and closed with 97% of payment instructions already converted. The CPMI’s 2025 monitoring survey found 79% of both RTGS and fast payment systems have implemented ISO 20022 or have concrete plans to by 2027. According to Axis Intelligence Research, the Harmonisation Alignment Gap (HAG) between that adoption figure and actual alignment with the CPMI data requirements stands at 63.5 percentage points across 137 surveyed systems, and 16.2 months remain to the end-2027 deadline.
Key Findings
- According to Axis Intelligence Research, the Harmonisation Alignment Gap (HAG) reads 66.0 percentage points for RTGS systems and 60.0 points for fast payment systems, meaning migration to ISO 20022 is running roughly five times ahead of alignment with the data rules that determine whether the format delivers anything.
- According to Axis Intelligence Research, only 21.2 of 137 surveyed payment systems (15.5%) are currently aligned with the CPMI harmonisation requirements, while 29.4 systems (21.5%) report no plans and are not considering alignment.
- Swift reported 97% of financial institution payment instructions in ISO 20022 on 24 November 2025, with a conversion service running to translate the residual MT traffic, per its 25 November 2025 completion statement.
- According to Axis Intelligence Research, the Mandate Enforcement Delay (MED) for CHAPS reaches 66.4 months: the system migrated on 19 June 2023, and the Bank of England has confirmed it will not reject payments solely for incomplete enhanced data before end-2028 at the earliest.
- The Bank of England rescinded its November 2027 structured remittance mandate for CHAPS in July 2026 and has published no replacement date, making it the first major enhanced-data requirement withdrawn after being set.
How Many Payment Systems Have Migrated to ISO 20022?
The 2025 CPMI cross-border payments monitoring survey found that 79% of real-time gross settlement systems and 79% of fast payment systems have either implemented ISO 20022 or hold concrete plans to do so by 2027, according to the CPMI’s updated harmonisation report published 26 February 2026. The survey covers 80 RTGS systems and 57 fast payment systems.
That headline hides the more useful number. Adoption of the format and alignment with the rules governing how the format is populated are separate questions, and the survey asks both.
ISO 20022 Adoption Versus Harmonisation Alignment, 2025 Survey
| Status | Fast payment systems | RTGS systems | Source |
|---|---|---|---|
| Implemented ISO 20022 or concrete plans by 2027 | 79% | 79% | CPMI, February 2026 |
| Already aligned with harmonisation requirements | 19% | 13% | CPMI, February 2026 |
| Plans to align | 12% | 25% | CPMI, February 2026 |
| Considering, no firm plans | 44% | 43% | CPMI, February 2026 |
| No plans and not considering | 25% | 19% | Axis Intelligence Research, from CPMI survey shares |
| Systems in scope | 57 | 80 | CPMI, February 2026 |
The CPMI notes that the count of aligned RTGS systems more than doubled since the 2024 survey round, driven primarily by seven systems executing plans they had already stated. Fast payment systems and RTGS systems in the Middle East and Africa were the most likely to say they would implement the requirements.
Sarah Davis: Read the two rows together and the shape of the problem is obvious. Nearly four in five systems are on the format; barely one in seven has committed to using it the way the cross-border chain needs it used. That is not hypocrisy, it is sequencing: a migration is a dated infrastructure project with a go-live and a board sponsor, while harmonisation is a data-quality discipline with no cutover and no ribbon to cut. The industry is very good at the first thing and historically poor at the second. What makes this cycle different is that the payoff everyone underwrote in the business case sits entirely on the second side of that line. Structured party data is what lifts straight-through processing rates and cuts false positives on sanctions screening. Nobody gets that from having migrated. They get it from having harmonised, and from their counterparties having harmonised too.
What Is the Harmonisation Alignment Gap (HAG)?
HAG stands for Harmonisation Alignment Gap. It measures the distance, in percentage points, between the share of payment systems that have adopted or committed to ISO 20022 and the share that are actually aligned with the CPMI’s harmonised data requirements. It exists because “migrated to ISO 20022” is the number the industry quotes and the number that does not predict any of the outcomes the G20 roadmap targets.
Formula
HAG = ISO 20022 adoption or committed-adoption share (%)
- Harmonisation-requirement alignment share (%)
Inputs: the 2025 CPMI cross-border payments monitoring survey, covering 57 fast payment systems and 80 RTGS systems, as published in the CPMI’s February 2026 updated report.
Worked example, RTGS systems
HAG_RTGS = 79.0 - 13.0
= 66.0 percentage points
The network-level reading weights by system count rather than averaging the two segment readings. Aligned systems total 57 x 19% + 80 x 13% = 21.2 of 137 systems, or 15.5%. So HAG_network = 79.0 – 15.5 = 63.5 percentage points.
HAG Readings by System Type, 2025
| Segment | Adoption or committed (%) | Aligned (%) | HAG (percentage points) | Source |
|---|---|---|---|---|
| RTGS systems | 79.0 | 13.0 | 66.0 | Axis Intelligence Research, from CPMI 2025 survey |
| Fast payment systems | 79.0 | 19.0 | 60.0 | Axis Intelligence Research, from CPMI 2025 survey |
| All surveyed systems | 79.0 | 15.5 | 63.5 | Axis Intelligence Research, from CPMI 2025 survey |
Interpretation. A HAG near zero means a segment’s stated migration converts into usable, interoperable data. A wide HAG means the format is in place and the benefit is not. The 6.0-point spread between fast payment systems and RTGS systems is the more interesting detail: instant payment rails, generally younger and built natively on ISO 20022, start closer to conformance than the high-value systems that mapped a legacy proprietary format onto the standard. According to Axis Intelligence Research, fast payment systems are 46% more likely to be aligned than RTGS systems on these survey shares, despite identical headline adoption.
Scope. HAG is computed on system counts, not on payment volume or value. A single aligned RTGS system carrying a quarter of global wholesale value counts the same as a small domestic one. The CPMI publishes the survey shares as rounded whole percentages, and the derived system counts inherit that rounding.
Where the Remaining Systems Sit
| Alignment status | Systems (count) | Share of 137 | Source |
|---|---|---|---|
| Already aligned | 21.2 | 15.5% | Axis Intelligence Research, from CPMI 2025 survey |
| Plans to align | 26.8 | 19.6% | Axis Intelligence Research, from CPMI 2025 survey |
| Considering, no firm plans | 59.5 | 43.4% | Axis Intelligence Research, from CPMI 2025 survey |
| No plans, not considering | 29.4 | 21.5% | Axis Intelligence Research, from CPMI 2025 survey |
The CPMI’s own reading is that if the considering and planning cohorts follow through, around two thirds of systems could align within a few years. Add the three cohorts and 78.5% of surveyed systems are somewhere on the path. According to Axis Intelligence Research, the decisive block is the 43.4% that is considering alignment without a firm plan, because with 16.2 months left to the end-2027 date, “considering” and “not aligned by the deadline” are converging into the same answer.
When Did ISO 20022 Become Mandatory for Cross-Border Payments?
Swift began the coexistence period between MT and ISO 20022 for cross-border payments and reporting in March 2023 and closed it on 22 November 2025. On 24 November, Swift observed 97% of payment instructions sent in ISO 20022, with a conversion service translating the remainder for a transitional period. The community chose the standard in 2018, making the elapsed decision-to-completion span seven years.
High-Value Payment System Migration Timeline
| System | Jurisdiction | ISO 20022 go-live | Migration style | Source |
|---|---|---|---|---|
| T2 | Euro area | 20 March 2023 | Big-bang, replaced TARGET2 | European Central Bank |
| CHAPS | United Kingdom | 19 June 2023 | Big-bang, enhanced fields enabled | Bank of England |
| CHIPS | United States | 8 April 2024 | Big-bang, no coexistence | The Clearing House |
| Fedwire Funds Service | United States | 14 July 2025 | Single-day implementation | Federal Reserve Financial Services |
| Swift CBPR+ | Cross-border | 22 November 2025 (coexistence end) | Phased, 32.1-month coexistence | Swift |
The Fedwire date carries a detail worth recording. The Board of Governors originally set 10 March 2025 as a single-day cutover, then rescheduled to 14 July 2025 after assessing customer readiness, per the Federal Reserve’s implementation FAQ. According to Axis Intelligence Research, that slip ran 126 days, and it made the largest USD wholesale rail the last of the five systems above to move onto the standard.
Annual Volume Now Running on Migrated Rails
| System | 2025 payment volume | Share of combined | Source |
|---|---|---|---|
| Fedwire Funds Service | 217,296,700 | 56.81% | Federal Reserve Financial Services; share Axis-calculated |
| T2 | 111,910,103 | 29.26% | European Central Bank; share Axis-calculated |
| CHAPS | 53,300,000 | 13.93% | Bank of England; share Axis-calculated |
| Combined | 382,506,803 | 100% | Axis Intelligence Research |
Value bases are reported in native currency and are not summed here: Fedwire originated USD 1,148,267,267 million in 2025, T2 settled EUR 492,859.0 billion, and CHAPS settled GBP 93.9 trillion. Axis Intelligence Research does not convert across three currencies at a single spot rate to manufacture a global total, because the three systems settle on different bases and a blended figure would carry a precision it does not have.
Sarah Davis: The volume table is the one to hand to anyone still describing ISO 20022 as an emerging standard. Three systems, 382.5 million wholesale payments a year, all natively on the format, and that excludes CHIPS, whose annual count is not published on the same basis. But look at what the mix tells you about testing exposure. Fedwire alone is 56.8% of that volume and it went live last, in July 2025, four months before Swift pulled the coexistence rug. The US high-value estate spent roughly a single quarter operating a new domestic format before the cross-border format became compulsory. Every other jurisdiction on that list had at least eighteen months of live running first. That sequencing is why the interesting operational data for 2026 will come out of North American correspondent desks rather than European ones.
How Many ISO 20022 Harmonisation Requirements Are There?
The CPMI sets 12 general harmonisation requirements applying to a core message set of nine ISO 20022 messages, with a further five customer-to-bank messages informing the harmonised data models. The requirements split into three groups.
The 12 CPMI Harmonisation Requirements
| # | Requirement | Group | Source |
|---|---|---|---|
| 1 | Use the appropriate message for a business function | Fundamental | CPMI, February 2026 |
| 2 | Use ISO externalised codes | Fundamental | CPMI, February 2026 |
| 3 | Restrict the character set to current market practice | Fundamental | CPMI, February 2026 |
| 4 | Use a common time convention | Fundamental | CPMI, February 2026 |
| 5 | Include a unique end-to-end reference (UETR) | Transparency | CPMI, February 2026 |
| 6 | Support transparency on amounts, conversions and charges | Transparency | CPMI, February 2026 |
| 7 | Include account identifiers to the extent possible | Data structuring | CPMI, February 2026 |
| 8 | Uniquely identify all financial institutions | Data structuring | CPMI, February 2026 |
| 9 | Identify all entities in a standardised, structured way | Data structuring | CPMI, February 2026 |
| 10 | Identify all persons in a standardised, structured way | Data structuring | CPMI, February 2026 |
| 11 | Provide structured minimum postal address information | Data structuring | CPMI, February 2026 |
| 12 | Transport customer remittance information end to end | Data structuring | CPMI, February 2026 |
The CPMI’s own impact mapping assigns direct speed effects to requirements 1 to 3, 5 and 7 to 11, and direct transparency effects to requirements 1, 2, 4 to 10 and 12. According to Axis Intelligence Research, that is 9 of 12 requirements bearing directly on speed and 10 of 12 on transparency, with the six data-structuring requirements appearing in both lists. The mechanism the CPMI identifies is unglamorous and consistent across all six: bulked, unstructured party and address data drives false positives in compliance screening, false positives drive manual intervention, and manual intervention is where the hours go.
The core message set covers credit transfers (pacs.008, pacs.009, pacs.002), payment returns (camt.056, camt.029, pacs.004) and payment investigations (pacs.028, camt.110, camt.111). Requirement 12 sets the remittance capacity that most corporate treasurers actually care about: a single occurrence of up to 140 characters unstructured, or multiple occurrences of structured remittance information up to 9,000 characters excluding XML tags. According to Axis Intelligence Research, that is a 64.3x capacity multiple over the unstructured field, and it is the single largest functional change between the MT world and the ISO 20022 world for reconciliation.
Sarah Davis: Requirement 6 is the one that will get litigated hardest, and the CPMI more or less says so. It asks for the instructed amount, any currency conversion applied, and any charges deducted by any institution in the chain, all carried unchanged end to end. The committee’s own note on implementation effort observes that complete transparency on amount and charges could have major implications for business models and competition. That is standards-body language for: this makes the FX margin visible, and the FX margin is where the money is. Nine of the twelve requirements are engineering. Requirement 6 is a pricing disclosure regime wearing an XML schema, and the alignment numbers for it will lag the others.
Why Did the Bank of England Drop Its Structured Remittance Mandate?
In September 2025 the Bank of England published its policy statement expanding mandatory ISO 20022 enhanced data in CHAPS from November 2027, having received over 20 consultation responses. In July 2026 it updated that statement to withdraw the structured remittance requirement for both pacs.008 and pacs.009 COV payments, citing industry feedback and a decision to pursue global alignment first. No replacement date has been published.
CHAPS Enhanced Data Mandate Timeline
| Date | Requirement | Status | Source |
|---|---|---|---|
| June 2023 | Enhanced fields enabled, not mandated | Delivered | Bank of England |
| May 2025 | Purpose codes for FI-to-FI and property payments; LEIs for FI-to-FI | In force | Bank of England |
| November 2027 | Purpose codes for all CHAPS payments | Confirmed | Bank of England |
| November 2027 | LEIs for pacs.004 returns and pacs.009 COV | Confirmed July 2026 | Bank of England |
| November 2027 | Structured remittance in pacs.008 and pacs.009 COV | Rescinded July 2026 | Bank of England |
| End-2028 at earliest | Rejection of payments for incomplete enhanced data | Not before this date | Bank of England |
According to Axis Intelligence Research, this is the first instance in the ISO 20022 programme of a major market infrastructure withdrawing an enhanced-data mandate it had already set, and it is a directly relevant precedent for every jurisdiction still drafting requirement 12 obligations. The Bank’s stated reason is interoperability rather than capability: mandating structured remittance domestically while Swift CBPR+ leaves the field optional produces a rail that demands data its own inbound flows do not carry.
The consultation record documents the mechanism. Respondents reported that a CBPR+ instruction arriving without required enhanced data would fall to a repair queue rather than straight-through process, so a domestic mandate not mirrored on the cross-border leg would show up as a decline in CHAPS STP rates rather than as better data. Swift is applying external code set validation in non-blocking mode under Standards Release 2025 and intends blocking mode under Standards Release 2026, at which point unregistered external codes are rejected outright.
The Purpose Code Problem
Consultation respondents raised the size of the code set as the principal risk to consistent usage. One respondent cited more than 300 purpose codes in the international ISO 20022 external code set and more than 100 on the UK recommended list; another put the existing UK list at 127 codes. Several reported that where end users face long lists they will choose the fastest option rather than the correct one, which degrades exactly the fraud-detection and analytics use cases that justified the mandate. The Bank’s response points to segmented code lists by customer cohort rather than a shorter master list.
Sarah Davis: There is a version of this story that reads as retreat, and it is the wrong reading. The Bank did not withdraw the remittance mandate because CHAPS participants cannot populate the field. It withdrew it because mandating high-quality data on one leg of a two-leg payment produces repair queues, not insight, and a repair queue is a cost with no offsetting revenue anywhere. That is a genuinely sophisticated call and it deserves more attention than it has had. It also sets the template for what happens next in every other jurisdiction: enhanced-data mandates only clear their own business case when the cross-border rulebook moves at the same time, which is precisely what the CPMI’s end-2027 date is for. Watch whether Swift makes purpose codes mandatory on CBPR+. Until it does, domestic mandates are one-legged and every operator now has a worked example of what that costs.
How Long Does It Take Before ISO 20022 Data Actually Becomes Enforceable?
MED stands for Mandate Enforcement Delay. It measures the elapsed months between a payment system’s ISO 20022 go-live and the date on which non-conforming data can actually cause a payment to be rejected or a participant to be in breach. It separates the technical migration date, which is what gets announced, from the date the data starts being reliable, which is what determines when downstream analytics, fraud models and screening improvements can be built on it.
Formula
MED = Months between ISO 20022 go-live date
and the date enhanced-data conformance becomes enforceable
Worked example, CHAPS
Go-live: 19 June 2023
Earliest rejection: 31 December 2028
MED_CHAPS = 2,022 days / 30.4375
= 66.4 months
MED Readings by System
| System | Go-live | Enforcement reference date | MED (months) | Source |
|---|---|---|---|---|
| CHAPS (rejection for enhanced data) | 19 June 2023 | End-2028 at earliest | 66.4 | Axis Intelligence Research, from Bank of England policy |
| CHAPS (all-payment purpose codes) | 19 June 2023 | 1 November 2027 | 52.4 | Axis Intelligence Research, from Bank of England policy |
| T2 (to CPMI harmonisation date) | 20 March 2023 | 31 December 2027 | 57.4 | Axis Intelligence Research, from CPMI and ECB |
| CHIPS (to CPMI harmonisation date) | 8 April 2024 | 31 December 2027 | 44.7 | Axis Intelligence Research, from CPMI and TCH |
| Fedwire (to CPMI harmonisation date) | 14 July 2025 | 31 December 2027 | 29.6 | Axis Intelligence Research, from CPMI and FRFS |
| CHAPS (first enhanced-data mandate) | 19 June 2023 | 1 May 2025 | 22.4 | Axis Intelligence Research, from Bank of England policy |
Interpretation. A short MED means data quality follows the format quickly and the analytics case matures early. A long MED means years of ISO 20022 messages that are structurally valid and substantively thin. The CHAPS reading of 66.4 months is the longest and also the best documented, because the Bank of England publishes its enforcement intentions explicitly where most operators do not. The Bank also commits to at least 18 months’ notice before expanding any requirement, which puts a hard floor under how fast any new obligation can arrive.
According to Axis Intelligence Research, the MED spread across these five systems runs from 29.6 to 66.4 months, a 2.24x range, and the direction is counterintuitive: the system that migrated last, Fedwire, faces the shortest runway to the harmonisation date, while the system that migrated first, T2, has the longest. Early movers bought optionality on data quality. Late movers have to do both jobs at once.
What Does ISO 20022 Data Show That MT Could Not?
The clearest published evidence sits in the Bank of England’s CHAPS statistics, which now break out payment types by ISO 20022 message rather than by the MT 202 catch-all that made the distinction impossible.
CHAPS Payment Composition, 2025
| Measure | Value | Source |
|---|---|---|
| Total payments | 53.3 million | Bank of England |
| Total value settled | GBP 93.9 trillion | Bank of England |
| Average daily volume | 210,482 | Bank of England |
| Average payment value | GBP 1.8 million | Bank of England |
| Median payment value | GBP 4,586 | Bank of England |
| Payments of GBP 1 million or less | 94% | Bank of England |
| pacs.009 (FI-to-FI) share of volume | 24% | Bank of England |
| pacs.009 share of value | 73% | Bank of England |
The mean-to-median ratio is the number that reframes CHAPS. An average of GBP 1.8 million against a median of GBP 4,586 says the system is overwhelmingly a retail-adjacent rail carrying a small population of very large wholesale transfers. According to Axis Intelligence Research, financial institution payments carry 3.04 times as much value per unit of volume as the CHAPS average, holding 73% of value on 24% of volume, and that split is only visible because pacs.009 separates what MT 202 combined.
The ECB documented the same effect on the other side of the Channel. When TARGET2 migrated to T2 in March 2023, average daily interbank payment turnover fell 9% against 2022, per the ECB TARGET Annual Report 2023, which the ECB attributed to reclassification rather than business change: MT 202 could not distinguish interbank payments from liquidity transfers, and ISO 20022 could, so a category that had been overstated for years corrected.
Sarah Davis: Two central banks, two years apart, both discovering that a chunk of what they had been reporting as interbank payment flow was liquidity management wearing the same message type. That is the actual first dividend of this programme, and it is a measurement dividend rather than a speed one. Anyone running a time series across the migration boundary on any of these systems needs to check whether their pre-2023 interbank series is comparable with the post-2023 one, because on T2 it demonstrably is not. The same caution applies to every domestic RTGS series that crosses its own go-live date. Our SWIFT payment statistics analysis covers the network-level traffic picture that sits above these domestic rails, and our cross-border payments statistics work covers the speed and cost outcomes the harmonisation programme is meant to move.
Methodology
Collection. Every figure in this dataset was retrieved from a primary source document during the production session on 25 August 2026. Eight source documents were used: the CPMI’s updated report on harmonised ISO 20022 data requirements (February 2026); Swift’s completion statement of 25 November 2025; the Federal Reserve Financial Services Fedwire Funds Service annual statistics and ISO 20022 implementation FAQ; the ECB’s TARGET Services Annual Report 2025 and TARGET Annual Report 2023; the Bank of England’s payment and settlement statistics and its September 2025 policy statement as updated in July 2026; and The Clearing House’s CHIPS migration statement of April 2024. No figure derives from secondary aggregation.
Formula: Harmonisation Alignment Gap (HAG).
HAG = adoption or committed-adoption share (%) - harmonisation alignment share (%)
Inputs are the 2025 CPMI cross-border payments monitoring survey shares across 57 fast payment systems and 80 RTGS systems. The network reading weights by system count: (57 x 0.19 + 80 x 0.13) / 137 = 15.5%, so 79.0 – 15.5 = 63.5 points.
Formula: Mandate Enforcement Delay (MED).
MED = (enforcement reference date - ISO 20022 go-live date) / 30.4375 days
The enforcement reference date is the earliest published date on which non-conforming enhanced data carries a consequence. Where an operator has published no such date, the CPMI end-2027 harmonisation date is used as the reference and the row is labelled accordingly.
Arithmetic. All derived figures were computed in Python from the source values and are reproducible from the accompanying CSV. Every calculated row carries a method note identifying its formula.
Disclosed source discrepancies. Two are worth stating rather than smoothing. First, the CPMI publishes survey shares as rounded whole percentages, so derived system counts (21.2 aligned, 29.4 with no plans) carry the rounding of their inputs and are presented to one decimal to make that visible rather than to imply precision. Second, purpose code counts differ between consultation respondents in the same Bank of England document, one citing more than 300 international codes and more than 100 UK codes, another citing 127 UK codes. Both are recorded as reported; Axis Intelligence Research does not reconcile them, because the disagreement between two participants about the size of the list they are being asked to implement is itself the finding.
What this reading does not capture. The CPMI survey is a census of payment system operators, not of the banks and payment service providers that populate the messages, so alignment at the infrastructure layer does not establish alignment at the participant layer. System counts are unweighted by volume or value. CHIPS annual volume is not published on a basis comparable with the other systems and is therefore excluded from the combined volume base rather than estimated.
About This Dataset
This dataset contains 97 observations covering ISO 20022 migration dates, adoption and harmonisation-alignment survey shares, annual volume and value on migrated high-value payment systems, enhanced-data mandate timelines, and two Axis-constructed metrics, spanning 2018 to 2028 with global, euro area, UK and US coverage.
Licence: CC BY 4.0
Citation line: Axis Intelligence Research, ISO 20022 Statistics 2026, 2026
How to Cite This Research
APA Axis Intelligence Research. (2026). ISO 20022 statistics 2026: Migration data, adoption rates and the 63.5-point harmonisation gap. Axis Intelligence. https://axis-intelligence.com/iso-20022-statistics/
MLA Axis Intelligence Research. “ISO 20022 Statistics 2026: Migration Data, Adoption Rates and the 63.5-Point Harmonisation Gap.” Axis Intelligence, 2026, axis-intelligence.com/iso-20022-statistics/.
Chicago Axis Intelligence Research. “ISO 20022 Statistics 2026: Migration Data, Adoption Rates and the 63.5-Point Harmonisation Gap.” Axis Intelligence. 2026. https://axis-intelligence.com/iso-20022-statistics/.
Frequently Asked Questions
Does migrating to ISO 20022 mean a payment system is CPMI-compliant?
No, and the survey data shows how wide the difference is. Migration means the system exchanges ISO 20022 messages; alignment means it populates them according to the CPMI’s 12 harmonisation requirements, including structured addresses, externalised codes, UETRs and end-to-end charge transparency. Adoption stands at 79% of RTGS and fast payment systems, alignment at 13% and 19% respectively. The CPMI is explicit that the harmonisation requirements are not regulatory requirements and that non-conforming payments are not rejected, though they face higher risk of delay and failure.
What happened to the MT messages that were not converted by 22 November 2025?
Swift put a conversion service in place to automatically translate residual MT instructions into ISO 20022 for a transitional period, so the 3% that had not migrated by the deadline continued to flow rather than failing. That conversion is a translation layer, not a data upgrade: an MT message converted to pacs.008 carries only the data the MT message held, which means unstructured party information stays unstructured. Converted traffic satisfies the format requirement and not the harmonisation requirements.
Why does the same payment system appear compliant on speed but not on transparency?
Because the CPMI’s requirements map to different G20 objectives and carry different implementation costs. Nine of the twelve requirements bear directly on speed and are largely engineering work: message selection, code lists, character sets, identifiers. Requirement 6, on amount, currency conversion and charge transparency, is the outlier. The CPMI’s own assessment is that complete transparency on amounts and charges could have major implications for business models and competition among institutions providing cross-border payment services, which is a different order of decision from a schema change.
Is the UETR the same thing as a transaction reference?
No. The transaction identification field is not unique across all cross-border payments and all involved entities, which is why the CPMI specifies the unique end-to-end transaction reference under RFC 4122 version 4 instead. Its practical advantage is that no central authority issues it: any party can generate one algorithmically, which is what makes end-to-end tracking work across a correspondent chain nobody administers centrally.
Why did the Bank of England withdraw structured remittance from the CHAPS mandate?
Interoperability rather than capability. Consultation respondents reported that a Swift CBPR+ instruction arriving without the enhanced data a domestic mandate requires would fall to a repair queue rather than straight-through process, so mandating the field on one leg of a two-leg payment would push CHAPS STP rates down without improving data quality on inbound flows. The Bank withdrew the requirement in July 2026 and stated it will engage with international peers on global alignment first. No replacement date has been set, and participants must still be able to receive structured remittance.
How many purpose codes are there, and why is that a problem?
Bank of England consultation respondents cited more than 300 codes in the international ISO 20022 external code set, with the UK recommended list variously put at more than 100 and at 127. The operational concern respondents raised is not the count itself but selection quality: presented with long lists, end users pick the fastest or the wrong option, and inaccurate purpose codes undermine the fraud-detection and payment-prioritisation cases that justified mandating the field. The Bank’s response is segmented lists by customer cohort rather than a shorter master list.
Can pre-migration and post-migration payment statistics be compared directly?
Not always, and the ECB documented why. TARGET2’s MT 202 message could not distinguish interbank payments from liquidity transfers, so both were counted in the same category. When T2 went live in March 2023 with ISO 20022, reported interbank turnover fell 9%, which the ECB attributed to reclassification rather than any change in participant behaviour. Any time series crossing a system’s ISO 20022 go-live date needs checking for the same effect before it is used.
What is the deadline after end-2027?
FATF’s revised Recommendation 16 on payment transparency, updated in June 2025, is expected to be fully implemented by end-2030, with FATF guidance anticipated by end-2026. The CPMI states its harmonised data model for persons and entities is identifier-neutral and aligned with those expectations, so institutions aligning to the harmonisation requirements by end-2027 are positioned for the FATF deadline. The CPMI will maintain the harmonisation requirements at least until end-2027 under the G20 programme.
Which US system migrated first, CHIPS or Fedwire?
CHIPS, by fifteen months. The Clearing House moved CHIPS on 8 April 2024 as a big-bang cutover with no coexistence period, releasing 555,345 payments worth USD 1.81 trillion on day one. The Fedwire Funds Service followed on 14 July 2025, having been rescheduled from an original date of 10 March 2025. CHIPS reports that 95% of its payments have a leg beginning or ending outside the United States, which is why it moved ahead of the domestic rail.
Did Fedwire payments get faster after the ISO 20022 migration?
No, and Federal Reserve Financial Services states so directly: the Fedwire Funds Service continues to process messages at the same speed as it did before 14 July 2025. That is the correct expectation for a domestic RTGS cutover. ISO 20022’s speed benefit in the CPMI framework comes from reduced manual intervention in compliance screening and exception handling along a multi-party cross-border chain, not from message throughput on a single system where straight-through processing was already the norm.
Related research:
SWIFT payment statistics | Cross-border payments statistics | Wire transfer statistics | ACH payments statistics | Same Day ACH statistics | Healthcare EFT statistics