Prevention of Tax Treaty Abuse – Third Peer Review Report on Treaty Shopping

4. Conclusion and next steps

57. The 2020 peer review shows that the MLI, which has been the main tool used to implement the minimum standard, has started to have a significant effect and is now strengthening the bilateral tax treaty network of jurisdictions that have ratified it.

58. This year’s peer review, however, reveals that the Action 6 minimum standard is being implemented unevenly and, in particular, that there is an important difference in the progress made on its implementation between jurisdictions that have ratified the MLI and other jurisdictions that have not.

59. Importantly, the peer review shows that jurisdictions that have not signed or ratified the MLI have still generally made no or very little progress in implementing the minimum standard. The 2020 peer review thus highlights that ratification of the MLI is an effective tool for jurisdictions that want to implement the minimum standard through the PPT.

60. The 2020 peer review also identifies gaps in the coverage of the MLI and the agreements of signatories or parties to the MLI that are neither covered under the MLI nor subject to bilateral renegotiations. In total, over 500 agreements concluded between jurisdictions that are members of the Inclusive Framework will not at this stage be modified by the MLI as one partner has either not listed the agreement under the MLI or signed the MLI.

61. Finally, the 2020 peer review identifies jurisdictions that made no or very little progress in the implementation of the minimum standard to ensure those jurisdictions can be provided with support when needed. In general, those jurisdictions have not signed the MLI or implemented anti-treaty-shopping measures in their treaties.

62. The 2020 peer review and the implementation issues identified provide useful insights for the ongoing review of the peer review methodology.

Next steps for certain members of the Inclusive Framework

63. This year’s peer review provides additional information on some jurisdictions’ progress towards the implementation of the minimum standard. The information provided could help jurisdictions in identifying the immediate next steps to take towards the implementation of the minimum standard. Those could include ratifying the MLI for signatories that had not done so, closing gaps in the MLI coverage or establishing a negotiation plan for the implementation of the minimum standard.

Next steps for the Inclusive Framework

64. The implementation of the minimum standard will continue to be monitored and, as set out in the Peer Review Documents, the next peer review exercise will be launched in the first half of 2021.

65. As noted at paragraph 14 of the Peer Review Documents, the methodology for the review of the implementation of the minimum standard would be reviewed in light of the experience in conducting the first three peer reviews conducted in 2018, 2019 and 2020.

66. At the time this report was written, the review was being discussed by Working Party 1 and it is expected that the 2021 peer review will be conducted under a revised peer review methodology, which the Inclusive Framework will have an opportunity to approve in the first part of 2021.

Background on the Action 6 Minimum Standard and Peer Review

Context of the peer review

67. Over the last decades, bilateral tax agreements, concluded by nearly every jurisdiction in the world, have served to prevent harmful double taxation and remove obstacles to cross-border trade in goods and services, and movements of capital, technology and persons. This extensive network of tax agreements has, however, also given rise to so-called “treaty-shopping” arrangements.

68. As set out in the Action 6 Final Report, treaty shopping typically involves the attempt by a person to indirectly access the benefits of a tax agreement between two jurisdictions without being a resident of one of those jurisdictions.1

69. Treaty shopping is undesirable for several reasons, including:

• Treaty benefits negotiated between the parties to an agreement are economically extended to residents of a third jurisdiction in a way the parties did not intend. The principle of reciprocity is therefore breached and the balance of concessions that the parties make is altered;

• Income may escape taxation altogether or be subject to inadequate taxation in a way the parties did not intend; and

• The jurisdiction of residence of the ultimate income beneficiary has less incentive to enter into a tax agreement with the jurisdiction of source, because residents of the jurisdiction of residence can indirectly receive treaty benefits from the jurisdiction of source without the need for the jurisdiction of residence to provide reciprocal benefits.

Some previous attempts to tackle treaty shopping

70. Concerns about treaty shopping are not new. For example, in 1977, the concept of “beneficial owner” was introduced into the dividends, interest, and royalties articles of the OECD Model Tax Convention to clarify the meaning of the words “paid to”, and deal with simple treaty-shopping situations where income is paid to an intermediary resident of a treaty country who is not treated as the owner of that income for tax purposes (such as an agent or nominee).2

71. In 1977, the Commentary on Article 1 of the OECD Model Tax Convention was also updated to include a section on the improper use of tax agreements.3 In 1986, the Committee on Fiscal Affairs (CFA) published two reports: Double Taxation and the Use of Base Companies and Double Taxation and the Use of Conduit Companies. In 2002, the Committee published the report, Restricting the Entitlement to Treaty Benefits. The Commentary on Article 1 of the OECD Model Tax Convention was expanded on several occasions, notably in 2003, with the inclusion of sample provisions that countries could use to counter treaty shopping.

72. A review of jurisdictions’ practices shows that they have tried to address treaty shopping in the past and have used different approaches to do so. Some have relied on specific anti-abuse rules based on the legal nature, ownership, and general activities of residents of a jurisdiction party to a tax agreement.4 Others have favoured a general anti-abuse rule based on the purpose of transactions or arrangements.

BEPS and treaty shopping

73. The BEPS Action Plan, developed by the CFA and endorsed by the G20 Leaders in September 2013, identified 15 actions to address base erosion and profit shifting (BEPS). It identified treaty abuse, and in particular treaty shopping, as one of the most important sources of BEPS concerns.

74. Action 6 (Prevent Treaty Abuse) of the BEPS Action Plan called for the development of treaty provisions to prevent the granting of treaty benefits in inappropriate circumstances. In parallel, Action 15 of the BEPS Action Plan called for an analysis of the possible development of a multilateral instrument “to enable jurisdictions that wish to do so to implement measures developed in the course of the work on BEPS and amend bilateral tax treaties”.

75. After two years of work, the CFA, in which OECD and G20 countries work on an equal footing, produced the final BEPS Package,5 which was endorsed by the OECD Council and the G20 Leaders in November 2015.

76.  Jurisdictions agreed that four of the BEPS measures would be minimum standards that participating jurisdictions would commit to implement. The Action 6 Report sets out one of these minimum standards. The Action 6 minimum standard requires jurisdictions to commit to include in their tax treaties provisions dealing with treaty shopping to ensure a minimum level of protection against treaty abuse.

The Action 6 minimum standard

77. The minimum standard on treaty shopping requires jurisdictions to include two components in their tax agreements: an express statement on non-taxation (generally in the preamble) and one of three methods of addressing treaty shopping.

78. The minimum standard does not provide how these two components should be implemented (i.e. through the MLI or amending instruments). It recognises, however, that these provisions need to be agreed bilaterally and that a jurisdiction will be required to implement the minimum standard when requested to do so by another member of the Inclusive Framework.

The express statement

79. As set out in paragraphs 22 and 23 of the Final Report on Action 6, jurisdictions have agreed to include in their tax agreements an express statement that their common intention is to eliminate double taxation without creating opportunities for non-taxation or reduced taxation through tax evasion or avoidance, including through treaty-shopping arrangements. The following provision now appears in the 2017 OECD Model Tax Convention: “Intending to conclude a Convention for the elimination of double taxation with respect to taxes on income and on capital without creating opportunities for non- taxation or reduced taxation through tax evasion or avoidance (including through treaty-shopping arrangements aimed at obtaining reliefs provided in this Convention for the indirect benefit of residents of third States).”

Three methods of addressing treaty shopping

80. Jurisdictions have also committed to implement that “common intention” through the inclusion of treaty provisions in one of the following three forms:

i. a principal purpose test (PPT) equivalent to paragraph 9 of Article 29 of the 2017 OECD Model Tax Convention, together with either a simplified or a detailed version of the limitation on benefits (LOB) rule that appears in paragraphs 1 to 7 of the 2017 OECD Model Tax Convention; or

ii. the PPT alone; or

iii. a detailed version of the LOB rule together with a mechanism (such as a treaty rule that might take the form of a PPT rule restricted to conduit arrangements, or domestic anti-abuse rules or judicial doctrines that would achieve a similar result) that would deal with conduit arrangements not already dealt with in tax treaties.

The obligation to implement the minimum standard

81. The Action 6 Report recognised that “some flexibility in the implementation of the Action 6 minimum standard [would be] required, as these provisions need to be adapted to each country’s specificities and to the circumstances of the negotiation of bilateral conventions.” In particular:

• a jurisdiction is required to implement the minimum standard in a treaty only if asked to do so by another member of the Inclusive Framework;

• its form (which of the three options used) has to be agreed (a solution cannot be imposed); and

• the commitment to adopt in bilateral treaties measures that implement the minimum standard should not be interpreted as a commitment to conclude new treaties or amend existing treaties within a specified period of time.

82. The Action 6 Report itself recognised that since participation in the multilateral instrument (see next section) was not mandatory and because jurisdictions could have different preferences about how the minimum standard should be met, monitoring its implementation would be necessary.

83. In May 2017, the Inclusive Framework agreed the Terms of Reference for the peer review and its methodology, and decided that the methodology would be reviewed in 2020.6

The 2018 peer review

84. The first peer review was conducted in 2018 and covered the 116 jurisdictions that were members of the Inclusive Framework on 30 June 2018. The Peer Review Report, which was adopted by the Inclusive Framework in January 2019, was published on 14 February 2019.

85. The 2018 peer review revealed that, as the provisions of the MLI had not taken effect at the time of the first peer review, nearly all of the agreements reviewed for this report did not at that time comply with the minimum standard. Substantial progress had, however, been made in 2017 and 2018 towards its implementation and a large majority of members of the Inclusive Framework had begun to translate their commitment on treaty shopping into actions and were in the process of modifying their treaty networks.

86. In total, on 30 June 2018, the peer review showed that 82 jurisdictions had some agreements that were already compliant with the minimum standard or were subject to a complying instrument that would bring their agreements into compliance.7 The first peer review highlighted the effectiveness of the MLI in implementing the treaty-related BEPS measures. It was by far the preferred tool of members of the Inclusive Framework for implementing the minimum standard.

87. In the course of the first peer review, all concerns raised by jurisdictions on the implementation of the minimum standard in their agreements had been resolved when the 2018 report was approved by the Inclusive Framework and therefore no recommendation was made under the first peer review.

The 2019 peer review

88. The second peer review was conducted in 2019 and covered the 129 jurisdictions that were members of the Inclusive Framework on 30 June 2019. The peer review report, which was adopted by the Inclusive Framework in January 2020, was published on 24 March 2020.

89. The 2019 peer review revealed that, by 30 June 2019, 91 members of the Inclusive Framework had begun to update their bilateral treaty network and were implementing the minimum standard. The data compiled for this peer review demonstrated that the MLI had been the tool used by the vast majority of jurisdictions that had begun to implement the minimum standard.

90. By 30 June 2019, the MLI had already modified around 60 bilateral agreements. The MLI’s impact was expected to increase quickly as jurisdictions ratified it.

91. In the course of the second peer review, a jurisdiction had raised a concern with respect to the CARICOM Agreement, a multilateral agreement concluded by eleven jurisdictions, ten of which were members of the Inclusive Framework. The CARICOM Agreement had been concluded in 1994 to encourage regional trade and investment within the CARICOM, and contains several unusual features,8 not found in the OECD Model Tax Convention or UN Model Double Taxation Convention, which could lead to certain income flows escaping tax altogether. These departures from standard tax treaty provisions may have encouraged greater economic integration within the CARICOM at the time, but they may also have made the CARICOM Agreement more vulnerable to treaty shopping and other forms of abuse. Previous renegotiation attempts of the CARICOM Agreement had proven to be difficult.

Conduct of the 2020 peer review

92. The review started with a questionnaire sent to members of the Inclusive Framework in March 2020. The questionnaire was based on the 2018 Action 6 peer review questionnaire with a few modifications. In 2018, each jurisdiction was asked to list all of its comprehensive income tax agreements in force. Those lists indicated whether each agreement included the provisions of the minimum standard, i.e. a complying preamble and the necessary anti-abuse provision(s), whether it was subject to a “complying instrument” that would soon bring it into compliance, what that instrument was, and which of the three methods of meeting the minimum standard had been used.

93. In 2020, each jurisdiction was invited to update its 2019 questionnaire taking into consideration any new agreements that had entered into force by adding them to their list of agreements. Jurisdictions that joined the Inclusive Framework after 30 June 2019, were asked to complete the original questionnaire.

94. Each jurisdiction was also asked to answer additional questions on the ratification of complying instruments and issues described in Sections D and E of the Peer Review Documents on difficulties encountered in getting agreement from another jurisdiction to implement the minimum standard. Jurisdictions were also free to add any further comments. The list of the 137 jurisdictions that were subject to the peer review and full details by jurisdiction are contained in Chapter 5.

95. The Secretariat analysed jurisdictions’ responses to verify and reconcile any divergent information and produced a first draft of this report.

Notes

← 1.  See paragraph 17 of the BEPS Action 6 Final Report (2015). As the Report also notes, cases where a resident of the Contracting State in which income originates seeks to obtain treaty benefits (e.g. through a transfer of residence to the other Contracting State or through the use of an entity established in that other State) could also be considered a form of treaty shopping.

← 2.  See paragraph 2 of Articles 10 and 11, and paragraph 1 of Article 12 of the OECD Model Tax Convention.

← 3.  See paragraphs 7-10 of the Commentary on Article 1 of the 1977 Model Tax Convention.

← 4.  “Limitation on benefits” provisions commonly found in treaties concluded by the United States are the best-known example.

← 5.  In October 2015, the CFA, including OECD and G20 countries working on an equal footing, produced the Final BEPS Package, in the form of reports on each of the 15 actions accompanied by an Explanatory Statement. The Final BEPS Package gives countries and economies the tools they need to ensure that profits are taxed where economic activities generating the profits are performed and where value is created, while at the same time giving businesses greater certainty by reducing disputes over the application of international tax rules and standardising compliance requirements.

← 6.  The review of the peer review methodology was being conducted at the time this report was written, in light of the experience in conducting the peer reviews.

← 7.  A further seven jurisdictions had no comprehensive tax agreements and were outside the scope of this exercise.

← 8.  The CARICOM Agreement provides for an almost exclusive source-based taxation of all income, gains and profits. Some income – for instance dividends – are also entirely exempted from tax under the CARICOM Agreement.