This chapter presents the key outcomes and aggregate results for the first 25 individual jurisdiction reports adopted by the Global Forum in 2025. It presents an analysis of the in-box recommendations examined for the 25 reports, a summary of findings on two areas where jurisdictions face particular difficulties in addressing recommendations, and an overview of EOIR in practice. It includes a summary of the peer input received so far in the enhanced monitoring process and presents the outcomes of the peer input analysed in the context of the first 25 reports.
Enhanced Monitoring Report on the Implementation of the Standard on Transparency and Exchange of Information on Request 2025
Abstract
Analysis of in-box recommendations
The 25 monitored jurisdictions reported actions taken in respect of a total of 125 in-box recommendations. Of these, 45 recommendations related to aspects of the jurisdictions’ legal and regulatory framework (“framework” recommendations), and 80 related to the practical implementation of the frameworks (“practice” recommendations).
Across the 25 reports, 36 in-box recommendations are “considered addressed in the context of the monitoring process (subject to a peer review)” or “considered provisionally addressed in the context of the monitoring process, subject to detailed validation” – together referred to as “considered (provisionally) addressed”. Both statuses convey a good degree of satisfaction that the reported actions address a recommendation. Further, 67 in-box recommendations are categorised as “in the process of being addressed” and 18 are assigned the status of “has not been addressed”. In addition, four]in-box recommendations are considered “no longer sufficiently material” (for the meaning of these statuses, refer to the discussion on “Statuses determined and actions required” under Chapter 1, section on “Procedure followed”, “Step 5 PRMG decisions”).
Figure 2.1 presents an element-wise and a type-wise (framework or practice) breakdown of the statuses determined for in-box recommendations analysed in the 25 reports.
Source: Global Forum Secretariat
In relation to the 36 recommendations that are “considered (provisionally) addressed”, 15 are recommendations on the jurisdictions’ legal and regulatory frameworks (framework recommendations), while 21 pertain to implementation aspects. Out of the 25 jurisdictions, 16 have at least one of their recommendations considered (provisionally) addressed.1 These addressed recommendations relate to a range of issues across the standard. In respect of framework recommendations, notable progress by jurisdictions is seen through the amendments of their legal and regulatory frameworks to address gaps on availability of ownership information. However, a legal change must be followed up with a demonstration of suitable implementation. In this regard, while 15 framework recommendations have been accepted as (provisionally) addressed, 7 new recommendations on the implementation of the amended framework have been issued, all of which are in respect of the availability of ownership information.
In respect of recommendations on implementation, one key area where progress has been made is effective exchange of information for which nine recommendations across seven jurisdictions are considered (provisionally) addressed. Each of these jurisdictions has taken measures to improve the timeliness of responding to their requests, including putting in place improved monitoring and tracking of incoming requests, providing trainings to staff and/or ensuring adequate staffing for the effective handling of requests.2
Sixty-seven recommendations (or 54%) are determined to be “in the process of being addressed”. Although this status means that the recommendations are not yet considered addressed, for almost all such recommendations, material progress has been made by the jurisdictions, even though some further work remains to be done or to be demonstrated to fully address the recommendation. For such recommendations, jurisdictions have been asked to take suitable actions and to report further progress on the implementation of these recommendations in their next self-assessments. Of the 25 jurisdictions, 21 have at least one recommendation considered as being “in the process of being addressed”.
Eighteen recommendations are each assigned the status of “has not been addressed”. Of the 25 jurisdictions, 7 jurisdictions have this status for at least one of their recommendations.3 Following the methodology (point 2 of the table at paragraph 60 of the methodology for enhanced monitoring), where more than three years have elapsed since the issuance of the recommendation and material progress4 has not been demonstrated, the jurisdiction must submit a schedule for completion of actions to address the recommendation and provide an annual update to the PRMG. Out of these seven jurisdictions, four have been asked to undertake such a follow-up and start reporting back from 31 March 2026. In addition, one jurisdiction with a recommendation “in the process of being addressed” has been similarly asked to undertake such a follow-up.5
Four recommendations have been ascertained to be “not sufficiently material” and no further reporting is required. Three of them pertain to monitoring that attorney-client privilege does not unduly affect the powers of the competent authority to access and exchange information. The jurisdictions concerned were able to demonstrate that there is no indication of any adverse consequence on EOIR in practice as a result; thus, although no specific legal or regulatory change has been made, there is no indication that this aspect has had or is likely to pose any difficulties to effective EOIR in practice. One recommendation was for the jurisdiction to ratify all signed treaties and, since its review, the jurisdiction signed and ratified the Multilateral Convention, which addressed the substance of the recommendation, as now the jurisdiction and its partners can exchange information in accordance with the standard.
Steady progress on the availability of beneficial ownership information
The defining feature of the revision of the Terms of Reference in relation to the EOIR standard in 2015 was the introduction of the requirement to ensure availability of beneficial ownership information for legal entities and arrangements, as well as for bank accounts. In the second round of reviews this new requirement has accordingly been one of the key focus areas. It has led to recommendations for most jurisdictions, as they put in place systems to comply with the revised standard; of the 125 in-box recommendations analysed in the 25 reports, 57 (46%) pertain to the availability of beneficial ownership information.
The challenges faced by jurisdictions include both the putting in place of the necessary legal requirements for ensuring the availability of beneficial ownership information and the practical implementation and supervision of these requirements; of the 57 recommendations related to beneficial ownership information, 27 recommendations pertain to the legal and regulatory framework and 30 recommendations pertain to implementation in practice. As banks and financial institutions in most jurisdictions were already familiar with the concept of beneficial owner and had already been implementing anti-money laundering requirements related to beneficial ownership information, only a small proportion of recommendations (8) relate to availability of beneficial ownership information on bank accounts (Element A.3), and the bulk of the recommendations (49) relate to the availability of beneficial ownership information of relevant legal entities and arrangements (Element A.1).
While jurisdictions have generally been making progress on addressing these recommendations, demonstrating effectiveness in practice takes time. In this regard, addressing issues in legal and regulatory frameworks outpace demonstrated improvements in practical implementation. Out of the 57 recommendations related to beneficial ownership, 11 recommendations relating mainly to legal and regulatory frameworks are “considered (provisionally) addressed” and 37 recommendations, mostly relating to practical implementation, remain “in the process of being addressed”. Finally, nine recommendations have each been determined as “has not been addressed”.
Some common trends on beneficial ownership information recommendations are noticeable.
Jurisdictions have progressively adopted a multi-pronged approach to ensure the availability of beneficial ownership information by putting in place more than one source of such information. For instance, obligations on entities themselves to maintain adequate, accurate and up-to-date beneficial ownership information, and/or the introduction of beneficial ownership registers to supplement the beneficial ownership requirements under the anti-money laundering framework.
Jurisdictions are demonstrating an increased understanding of the complexities of ensuring the availability of accurate beneficial ownership information and the associated challenges. The experience of jurisdictions shows that the legal framework for beneficial ownership registers takes time to mature and challenges in practical implementation often require legislative solutions.
The practical implementation of beneficial ownership registers follows an evolutionary trajectory – this begins with outreach and enforcement measures targeted at the persons with obligations with respect to the beneficial ownership register, followed by data cleaning, matching and verification and, finally, risk-based supervision.
Austria established its Beneficial Ownership Register in 2018 and has, over time, ensured that it covers 96% of all legal entities and arrangements in Austria. The Beneficial Ownership Register has been interconnected with other registers, allowing for the pre-population of beneficial ownership information for entities with a simple ownership chain, as well as the carrying out of monitoring activities for all entities through the continuous synchronisation of the data between registers. The accuracy of the beneficial ownership information is further ensured by a requirement for entities to verify beneficial ownership information at least once a year and to report to the register any known changes to their beneficial owners within four weeks. AML-obliged entities and authorities, having access to the Beneficial Ownership Register, are required to report any discrepancies identified. The Registry Authority monitors the quality of submitted information through an automated system that identifies samples (random, risk-based or ad-hoc) for monthly reviews. It then assigns each beneficial ownership report with a risk level indicating the likelihood that reported information may be incorrect. The Registry Authority follows up on issues identified with measures that depend on the gravity of findings. In addition, analytical tools have been implemented to identify potential companies or entities that may be used for the circumvention of European sanctions. Austria has implemented a dissuasive and proportionate penalties framework. The implemented monitoring and enforcement framework is subject to regular reviews to include any new risks.
Of the 20 monitored jurisdictions that received an in-box recommendation on implementing their beneficial ownership registers, some have reported high levels of success in populating the beneficial ownership registers.6 Discrepancy reporting by AML-obliged persons has emerged as an important tool to verify the information reported to beneficial ownership registers.7 For most jurisdictions, more work is still needed to ensure the quality of the information in the beneficial ownership registers. Hence, the reports of some jurisdictions acknowledge the progress on the implementation of their beneficial ownership registers and indicate that they should take further supervisory steps and report the results to demonstrate the effectiveness of their efforts to ensure the quality of information contained in the registers in due course.8
Availability of information on inactive entities
In EOIR peer review reports, inactive entities have been identified as posing specific issues in respect of the availability of ownership and accounting information, particularly where the reports noted a high volume of such entities and/or an inability of the jurisdiction to accurately estimate their numbers. Across jurisdictions and reports, inactive entities are defined and understood with some variation and no homogenous definition exists. A common definition is where entities have been assigned a status of “inactive” (or similar) by the authorities after their lack of economic activity has been verified. In such cases, there is a clear indication of the number of such entities. This usually requires coordination between tax authorities and the commercial registry, and the harmonisation of their databases. In other jurisdictions, inactive entities are not defined specifically but are understood to mean persistently non-compliant entities that fail to meet their filing requirements under tax law or commercial laws. Usually, the tax authorities and commercial registry assign their own statuses to entities they oversee based on the non-compliance they identify. This often leads to differences in the number of inactive entities between the databases of tax authorities and commercial registries.
The risk in respect of inactive entities is their continued legal personality that can allow them to carry on economic activities or to hold assets inside or outside of their jurisdiction of incorporation. Experience during peer reviews shows jurisdictions making EOI requests seeking information on entities which were or have been considered inactive by the requested jurisdiction. In such instances, requested jurisdictions have faced difficulties in providing the relevant legal and beneficial ownership information, as well as basic accounting information.
For the 25 reports, recommendations on the availability of ownership and accounting information in respect of inactive entities were present in 13 recommendations made to 7 jurisdictions.9 Following the analysis, no recommendation is “considered (provisionally) addressed”. Four recommendations issued to two jurisdictions (each with a recommendation in respect of ownership and accounting information) are each determined as “has not been addressed”. The remaining recommendations are “in the process of being addressed”.
Some facts noted so far are below:
Jurisdictions are at different stages in addressing recommendations pertaining to inactive entities.
A first step is often introducing a clear definition of inactive entities and establishing how many entities are inactive according to the definition. Jurisdictions are at different stages in this regard. Some jurisdictions have taken preliminary steps to coordinate between the relevant supervisory authorities to identify inactive entities, while others have taken steps to put in place and improve their monitoring tools. The more advanced jurisdictions have a more complete oversight over the number of inactive entities.
Jurisdictions have often relied on their existing surveillance mechanisms to monitor domestic economic activity. While this measure is helpful, it does not by itself mitigate the concerns around economic activities that might be carried on overseas or assets held abroad.
Some jurisdictions have adopted the approach of registry clean-up by the compulsory dissolution of inactive entities. This is usually the case where it is easy to identify that the entities do not hold assets that would require liquidation. Nevertheless, such jurisdictions should ensure the availability of information for at least five years following the dissolution of such entities.
The continued maintenance of legal personality by inactive entities continues to pose risks. There are variations across jurisdictions in their approach to clean-up the commercial registries by compulsorily dissolving inactive entities to extinguish their legal personality. Such actions are not yet legally feasible in some jurisdictions and entail costs and time around closure audits and liquidations.
Overall, jurisdictions continue to face challenges in demonstrating the sufficiency of actions to address recommendations pertaining to inactive entities.
EOIR experience
The EOIR experience of the monitored jurisdictions is analysed below through:
A discussion on the overall incoming and outgoing requests observed.
An analysis of the peer input received, including:
Overall trends
Observations specific to monitored jurisdictions
Incoming and outgoing requests
In total, over the 2023-2024 monitoring period, the 25 monitored jurisdictions received 11 682 requests for information and sent 17 401 such requests. While the statistics on timeliness of responding to incoming requests vary across jurisdictions, the following observations are made:
On average, 76% of incoming requests were responded to within 90 days and a further 14% within six months.
Improvement in the timeliness of providing responses to incoming requests resulted in the in-box recommendations on timeliness under Element C.5 for six jurisdictions being “considered (provisionally) addressed”.10
If a requested jurisdiction is unable to obtain and provide the information requested within 90 days of the receipt of a request, it should immediately inform the requesting partner, explaining the reason for its inability, the nature of the obstacles, or the reasons for its refusal, i.e. provide a status update on the request. Status updates were provided in 71% of all such cases across the 25 jurisdictions, with eight jurisdictions doing it systematically, i.e. always providing status updates where responses could not be provided within 90 days. The general trend seen is an improvement in this regard as compared to previous years, i.e. communication between authorities is improving. In-box recommendations on status updates under Element C.5 were considered (provisionally) addressed for 2 jurisdictions.11
Peer input – Overall trends
Substantial increase in overall peer input…
A key driver of progress to deliver effective EOIR in practice is the engagement of exchange partners to address issues, including through monitoring and peer review processes.
In the context of the enhanced monitoring process, a total of 802 pieces of peer input were received from 52 jurisdictions that related to all 82 jurisdictions covered under cycle 1. This is a very significant increase in the participation of Global Forum members in the provision of peer input on their EOI relationships. This demonstrates confidence in the enhanced monitoring process to deliver effective EOIR. Since 2021, when the submission of peer input was first introduced as part of the prior annual follow-up process, the submission of peer input has increased more than five times with the submission of peer input growing from 150 peer inputs in 2021, to 802 in 2025. What is most encouraging is the very high levels of positive input.
Distribution of peer input received from jurisdictions (2021-2025): Breakdown by year and feedback
Source: Global Forum Secretariat
…with more than three-quarters noting general level of satisfaction.
Most of the peer input on the monitored jurisdictions was positive, reflecting general satisfaction with the EOIR relationships. The proportion of peer input that was positive rose from 62% in 2021 to 79% in 2025. This trend is a testament of the positive EOIR relationships amongst Global Forum members.
Heightened interest as more members provide peer input, including for the first time…
Under the 2025 enhanced monitoring process, 16 members provided peer input for the first time (i.e. these members had not provided peer input under the earlier annual follow-up process). More generally, 35 members provided peer input on at least 5 EOI partners and 19 members provided peer input on more than 10 EOI partners, with some members providing input on a very significant number of their partners. This reflects a strong interest in the enhanced monitoring process and the confidence in its ability to deliver results.
Note: The bars illustrate the 52 Global Forum members that provided input on the monitored jurisdictions and the number at the outer end of each bar indicates the number of monitored jurisdictions on which they provided peer input.
Source: Global Forum Secretariat
…with input provided systematically across all cycle 1 jurisdictions
In their peer input, peers indicated the significance of the jurisdiction reported upon relative to the peers’ overall EOIR experience. A fifth of all received peer input, i.e. 157 out of the 802 peer inputs was in respect of monitored jurisdictions that were among the top five EOI partners for the peer submitting the input. A further one-third (261 peer inputs) was in respect of what the peers considered standard partners, i.e. partners which may not have been among the top five in terms of volume, but with whom the peer had either a significant volume of exchange or with whom there were more than a few simple requests (sent or received). Thus, more than 50% of all peer input pertained to monitored jurisdictions where the volume or complexity of requests made them important EOIR partners. This reflects the strong interest by members to provide feedback on their EOIR relationships in such cases.
There was a substantial volume of peer input that was received where the peer did not have a significant volume of exchanges or, in some cases, no exchanges over the monitoring period. In some instances, the reporting peer wanted to raise matters such as difficulties in communication (something that is often noted amongst new EOI partners). Where such peer input did not raise a specific issue, it often still provided a degree of reassurance that the absence of requests between the reporting peer and the monitored jurisdiction was not due to any problems that the peer had encountered in the past or anticipated with respect to the jurisdiction that might have deterred it from making its requests.
Source: Global Forum Secretariat
Peer input – Observations specific to monitored jurisdictions
In respect of the 25 reports, 252 peer inputs from 42 peers were analysed. Across these reports, 73% of the peer input was positive and reflected general satisfaction. Fifteen per cent of the input reflected that there had not been any exchange of information between the reporting peer and the monitored jurisdiction and, hence, the EOIR relationship was untested during the monitoring period.
Some peer input raised matters indicating one or more areas of improvement for the monitored jurisdictions. Thirty one out of the 252 peer inputs (or 12%) raised at least one matter on which the response of the monitored jurisdiction was needed. Of the 25 jurisdictions, 14 were requested to provide their response on such matters.
Number of peer inputs examined in the initial 25 reports: Broken down by type of feedback
Source: Based on the analysis of peer input received by the Global Forum Secretariat
Based on the responses received from the jurisdictions, in respect of the 25 reports, 19 matters were identified to be bilateral matters that were resolved bilaterally, mainly on issues of communication, generally due to technical issues. This reflects that, upon receiving the peer input, monitored jurisdictions proactively resolved such bilateral matters.
Some of the issues raised by peers were already captured by recommendations issued to the monitored jurisdictions in their peer review, and which the jurisdictions reported were “in the process of being addressed”. Typically, these matters pertained to provision of status updates and the timeliness of providing responses. In respect of the 25 reports, there were three instances where the matters raised led to the detection of a systemic issue of compliance with the standard. These matters were presented to the assessment panel for its views, which were subsequently submitted to the PRMG for consideration.
In two cases, peer input on incoming requests received by the monitored jurisdictions suggested backsliding as the jurisdictions had a much better record of effective exchange of information at the time of their EOIR peer review reports. Both jurisdictions were issued recommendations in this regard.12 In the third instance, the peer input raised a matter in relation to outgoing requests of the monitored jurisdiction and the matter was not resolved bilaterally. The jurisdiction was recommended to ensure the provision of clarifications in a timely manner to its treaty partners.13
← 1. Andorra, Aruba, Austria, Bahrain, Belgium, Bermuda, Dominica, Dominican Republic, Greece, Isle of Man, Japan, Liechtenstein, Luxembourg, New Zealand, Norway, and Saudi Arabia.
← 2. See the reports of Aruba, Dominica, Dominican Republic, Greece, Liechtenstein, Luxembourg and Saudi Arabia.
← 3. Andorra, Argentina, Barbados, Dominica, Greece, Korea, and New Zealand.
← 4. In considering material progress on a recommendation, attention has been paid to the nature and extent of the reported actions. Ordinarily, actions in the direction of addressing a recommendation and that are not just initial or preliminary steps, have been considered material progress and jurisdictions have been asked to continue their work and report back in the next self-assessment. When actions reflect only preliminary steps or do not give sufficient comfort that they have been in the direction of addressing the underlying deficiencies, it has been determined that material progress has not taken place. For such recommendations, jurisdictions are subject to closer monitoring.
← 5. Andorra, Greece, Korea, and New Zealand for recommendations determined as “has not been addressed” and Japan for a recommendation determined as “in the process of being addressed”.
← 6. See the reports of Austria, Bahrain, Belgium, Norway, Luxembourg, Liechtenstein, San Marino and Singapore.
← 7. Belgium, Greece, Luxembourg, Liechtenstein, Norway and San Marino, have indicated requirements of discrepancy reporting as a means for ensuring accuracy of the information in their beneficial ownership registers.
← 8. See the reports of Belgium, Isle of Man, San Marino and Singapore.
← 9. See the reports of Andorra, Argentina, Barbados, Dominican Republic, Korea, Saudi Arabia and Spain.
← 10. Jurisdictions where recommendations pertaining to timeliness of responding to requests have been considered provisionally addressed are Aruba, Dominican Republic, Greece, Luxembourg, Liechtenstein and Saudi Arabia.
← 11. See reports of Aruba and Saudi Arabia.
← 12. Evidence of backsliding on timeliness was noted in the reports of Andorra and Bermuda.
← 13. For further analysis of the matter, see the report of France.