What private equity funds of the future must take into account

Governments establish tax policy (such as defining the tax base and the tax treatment of nonresident investment capital) and create rules to implement that policy. TAs are tasked with administering the applicable tax rules, including claims under tax treaties and government directives, guided by the policy choices that informed the creation of those rules.

This section discusses some theories of tax treaty rationale and operation, and then considers some of the specific treaty terms that implement those concepts in the context of private equity fund investment and particularly IHCs. This background and history lead to a discussion of the OECD’s efforts to address issues specific to the investment industry, first with CIVs (which are generally seen as sympathetic but conceptually knotty) and non-CIV funds (such as private equity). The discussion then briefly considers the same points for the EU PSD before turning to key features of claims administration for both treaties and the EU PSD.

a. Tax treaties

i. Theory

Tax treaties have several stated purposes, but the present discussion is concerned with only the prevention of double taxation. The theory for preventing double taxation is relatively straightforward. Traditionally, states accept the primacy of taxing claims based on source of income, but states also tax persons based on residence. If source jurisdiction and residence jurisdiction both claim the right to tax the same income at the same time, then the economic activity generating the income may be taxed twice, which can impact economic activity in the jurisdiction. Faced with the potential loss of economic activity, the relevant jurisdictions may agree to allocate taxing authority between them.

An intermediary or intermediate legal entity between residence jurisdiction and source jurisdiction complicates treaty application by raising questions regarding concurrent claims. By way of a brief example, a nominee ordinarily is not considered to have an economic claim on income, and the principal (and thus the principal’s residence jurisdiction) does, so only a treaty between the source jurisdiction and the principal's residence jurisdiction ordinarily would apply. Under the OECD framework, a legal entity considered fiscally transparent by an owner’s residence jurisdiction generally is treated as that owner’s nominee.²⁰ If, in contrast, the owner’s residence jurisdiction does not consider the entity fiscally transparent, then the owner’s residence jurisdiction generally does not have a concurrent taxing claim.²¹ This tends to preclude owner claims for source-jurisdiction treaty benefits, but it also raises the question whether the entity itself may be so eligible. Parsing such arrangements is at the heart of concerns over treaty shopping.

ii. Scope

As used by the OECD, “treaty shopping” generally refers to “arrangements through which a person who is not a resident of a Contracting State [that is party to a particular tax treaty] may attempt to obtain benefits that a tax treaty grants to a resident of that State.”²² Historically, the OECD has identified three policy concerns raised by treaty shopping. First, “treaty benefits negotiated between two States are economically extended to persons resident in a third State in a way unintended by the Contracting States; thus the principle of reciprocity is breached, and the balance of sacrifices incurred in the tax treaties by the contracting parties altered,” (i.e., concern about the incidence of benefits).²³ Second, ”income flowing internationally may be exempted from taxation altogether or be subject to taxation in a way unintended by the Contracting States. This situation is unacceptable because the granting by a country of treaty benefits is based, except in specific circumstances, on the fact that the respective income… at least falls under the normal tax regime of that State,” (i.e., concern about deferral of taxable income for an owner of the entity claiming treaty benefits).^23.1 Finally, “[t]he State of residence of the ultimate income beneficiary has little incentive to enter into a [tax] treaty with the State of source, because the residents of the State of residence can indirectly receive treaty benefits from the State of source without the need for the State of residence to provide reciprocal benefits.”^23.2 As applied to non-CIV funds, the BEPS Action 6 Final Report identifies incidence and deferral as the principal concerns with treaty shopping in non-CIV funds.²⁴

iii. Qualification

The conditions for treaty qualification under the OECD Model have evolved over time, but, in its current form, a claimant generally must satisfy three conditions: claimant must be a “resident” (within the meaning of the treaty) of the relevant jurisdiction, claimant must be the “beneficial owner” (again, within the meaning of the treaty) of a payment with respect to which benefits are claimed, and the overall arrangement must comport with applicable anti-abuse provisions.

This subsection briefly discusses the resident and beneficial owner concepts before turning to a more expansive consideration of anti-abuse provisions. Starting from the OECD’s recently established treaty anti-abuse baseline that includes both a specific anti-abuse rule (the LoB) and a more general anti-abuse standard (the PPT), the discussion considers their history and policy with an eye toward their importance for private equity.

A. Resident

Treaties are available to only the contracting states’ “residents,” generally defined as persons liable to tax under domestic law.²⁵ Residency of a legal entity for tax purposes typically is determined either by reference to its formation jurisdiction (e.g., as the United States does) or by reference to its situs of management and control (e.g., as most other jurisdictions do). These differences can give rise to dual residency or no residency, which can raise difficult tax policy questions.²⁶ Notably, “liable to tax” does not necessarily mean taxed in fact, but does mean taxable in principle, including exempt status.²⁷ Administratively, this is commonly evidenced by a tax residency certificate (TRC) issued by the relevant TA.²⁸ In contrast, fiscal transparency definitionally implies that liability to tax falls on the entity’s owners. Thus, a fiscally transparent entity may not be a resident for treaty purposes, notwithstanding that it may be resident for other local tax purposes (e.g., filing obligations arising from local activity).²⁹

B. Beneficial owner

The “beneficial ownership” concept has been part of the OECD Model since at least 1977.³⁰ The term went largely undefined in the OECD Model until 2002, when the commentary was modified to include an appeal to the notion of dominion and control, with agency as the counterexample.³¹ Failing to qualify as a beneficial owner entails loss of the treaty claim, typically on the theory that the subject entity is acting as an agent for the owner/ principal.

C. Anti-abuse

Overview. The BEPS project resulted in a minimum anti-abuse standard for inclusion in all OECD treaties: the common intention to eliminate double taxation without creating opportunities for nontaxation or reduced taxation through tax evasion or avoidance, including through treaty shopping arrangements.³² The OECD acknowledges that the minimum standard can be implemented in any of a variety of ways, but the most comprehensive approach combines a statement of intent that specifically abjures creating opportunities for treaty abuse, including treaty shopping, with a specific anti-abuse rule in the form of an LoB and a general anti-abuse standard in the form of a PPT.³³ Many jurisdictions apply a so-called substance standard to holding entities. The following discussion provides a brief overview of each of these in turn.

Principal purpose test. The amended preamble to the OECD 2017 Model provides that the Convention is intended in part to prevent double taxation without creating opportunities for nontaxation or reduced taxation through tax evasion or avoidance, including through treaty-shopping arrangements.³⁴ The PPT added as part of the effort to achieve this goal allows denial of a treaty benefit “if it is reasonable to conclude, having regard to all relevant facts and circumstances, that obtaining that benefit was one of the principal purposes” of the arrangement, “unless it is established that granting that benefit in these circumstances would be in accordance with the object and purpose of the relevant provisions of [the] Convention.”³⁵

Drafters often try to give some content to such a “facts and circumstances” standard by providing examples, and the OECD 2017 Model includes several.³⁶ Among these, several relating to non-CIV funds were developed after other examples had been settled and after rejecting as too specific a variety of examples submitted by industry in response to an OECD request. Drafts of these three non-CIV examples (two of which are potentially in point for private equity; the third concerns a securitization arrangement) were published for public comment and ultimately were incorporated into the OECD 2017 Model commentary essentially without material change.

In Example K, a single institutional investor that is regulated in its residence jurisdiction establishes a regional platform with a single holding entity in a treaty jurisdiction that holds multiple assets across multiple jurisdictions and employs “an experienced local management team to review investment recommendations.” The holding entity “pays tax” in its residence jurisdiction and has a more favorable treaty with a particular target jurisdiction than the investor’s residence jurisdiction does.

In Example M, a real estate fund establishes a regional platform with a single holding entity in a treaty jurisdiction. The holding entity manages all of the fund’s real estate assets, each of which it holds in a separate company in the relevant source jurisdiction in order to ring-fence liabilities and facilitate debt financing. All of the investors are equivalent beneficiaries.³⁷

Comments from the non-CIV industry on public drafts of these examples focused on a variety of desired changes, including allowing Example K’s holding entity to outsource activities; identifying Example K’s holding entity as “taxable” rather than actually taxed; relaxing Example M to include a non-CIV majority owned by equivalent beneficiaries (in the same way that, in Example D, a CIV majority owned by equivalent beneficiaries passes the PPT); and defining “platform” in Examples K and M to include the use solely of per-asset SPVs in the holding jurisdiction, rather than a single aggregating entity. Some have taken Examples K and M to ratify in principle the viability of an aggregating holding entity.³⁸ However, because the OECD made no material changes in response to comments, it is difficult for a typical private equity fund owned by less than 100% equivalent beneficiaries to take much, if anything, more from the examples than that relevant factors include both in-house substance at an IHC, and whether an IHC’s claimed treaty benefit exceeds the benefit an investor could claim on an unintermediated investment. With respect to the latter point, it should be noted that the multilateral instrument (MLI) developed to allow governments to quickly implement BEPS actions (see additional discussion in Appendix) includes optional discretionary facts-and-circumstances relief for PPT failure if benefits denied under a PPT “would have been granted … in the absence of the transaction or arrangement” (the term “equivalent beneficiary” is not mentioned), which, curiously few states have elected.³⁹

The immediate consequence of failing a PPT obviously is losing the treaty claim, but the theory of disqualification can have further implications. For example, anti-abuse rules often (but not always) apply by ignoring the arrangement considered to be abusive, in which case it may be possible to analyze the facts considered as enduring.⁴⁰

Substance. The concept of “substance” derives from the transfer pricing principle that the amount and location of income generally should align with the value and location of the various activities that generate it (e.g., as reflected in the BEPS Action Plan).⁴¹

A variety of tax contexts that are not necessarily about treaty qualification but share a similar analytical approach invoke substance as a guide. The OECD’s 1998 report on harmful tax competition noted that one indicator of a potentially harmful regime is the ability to access it with few or no substantial activities relating to the relevant income.⁴² That report recommended work on transfer pricing concepts to align income with activities, and, although the harmful tax project’s focus evolved away from substance and toward transparency and information exchange, BEPS Action 5 renewed the focus on substance.⁴³ Similarly, the BEPS Action on controlled foreign corporation within the meaning of section 957 (CFC) invokes substance as an analytical factor.⁴⁴

The EC’s own formal efforts on substance date at least to the formation of the Code of Conduct (Business Taxation) Group, which noted in its 1999 report to the ECOFIN Council the absence of economic substance in tax-driven holding companies.⁴⁵ More recently, the EC has proposed an anti-tax avoidance directive (ATAD 3 or UNSHELL) that defines presumptive substance standards applicable to all EU entities, with the stated goal of preventing EU resident entities from claiming benefits under an EU Directive or a tax treaty with an EU member.⁴⁶

Limitation of benefits. The United States originally developed LoBs to address concerns about treaty shopping.⁴⁷ Although BEPS Action 6 introduced both “simplified” and “detailed” LoBs to the OECD 2017 Model, the bilateral complexity of the detailed LoB led the OECD to include in the MLI only the simplified LoB, of which there has been almost no uptake as of this writing.⁴⁸

An LoB generally provides objective threshold conditions designed to ensure that treaty benefits go to only residents of the other contracting state and not to residents of third states that do not have a substantial business and tax nexus with the other contracting state.⁴⁹ The core objective test specifies both a minimum percentage ownership by persons resident in either of the treaty states (the ownership test), and a maximum portion of the entity’s income payable in deductible payments to persons not resident in either contracting jurisdiction (the base erosion test). A savings clause affords discretion to grant benefits where an arrangement fails the objective test but does not fail the PPT.

The scope of acceptable beneficiaries has expanded with time, in many cases reflecting categories deemed “bona fide” in the OECD’s 1986 report on conduit companies.⁵⁰ Among these categories, “derivative benefits” provisions allow a company that otherwise fails the core LoB to qualify if it is owned (either entirely or to some specified extent) by up to a specified number of equivalent beneficiaries, notwithstanding the heightened potential for deferral and incidence problems.⁵¹

EU nondiscrimination. LoB provisions have always raised concerns about the potential for impermissible discrimination under one or more of the “four freedoms” at the center of the EU.⁵² For example, under a baseline LoB, an EU company with owners resident in the company’s residence jurisdiction could receive better treatment than the same company if owned by persons resident elsewhere in the EU.⁵³ Derivative benefits provisions mitigate, but probably cannot eliminate, the possibility of discrimination.⁵⁴ In 2016, the European Commission ruled that the LoB in the Japan-Netherlands treaty infringes the “freedom of establishment” principle.⁵⁵ The Commission dropped the infringement claim in 2020 but presumably could take it up with the European Court of Justice in the future.⁵⁶ The OECD has alluded to EU nondiscrimination concerns in the LoB context and has noted them explicitly in the CFC context.⁵⁷

PPT, CIVs and proportional benefits. Despite EU nondiscrimination concerns, the OECD has done considerable work on treaty claims by CIVs. The essence of it is that a CIV may or may not qualify for a given treaty on the basis of the CIV’s particular legal characteristics and local laws, and source countries should have flexibility to grant benefits on the basis of their ownership, largely on the theory of investor equivalence.⁵⁸ For example, a rule could allow a CIV to claim residence jurisdiction treaty benefits if it is some minimum percentage owned by a combination of persons resident in the CIV’s residence jurisdiction and equivalent beneficiaries, or to the extent of its ownership by equivalent beneficiaries (proportional benefits).⁵⁹ Alternately, a rule could allow the CIV to claim benefits on behalf of its treaty-eligible owners (rather than in its own name) if the relevant information can be developed.⁶⁰ The policy concern expressed in response to these approaches remains that any rule referencing ownership “changes the bilateral nature of a treaty negotiation,” which is to say, raises deferral and incidence concerns.⁶¹ Deferral seemingly can be addressed by limiting eligible CIVs to those that make annual earnings distributions.⁶² Incidence appears to be mostly a problem of administration (i.e., developing and sharing the necessary information accurately and timely), as further discussed immediately below.

iv. Administration

The two primary approaches to implementing treaty-reduced rates involve either reduction at source (payer/withholding agent withholds and files relevant returns with the tax authority) or higher-rate withholding followed by an application for a refund. Claimants obviously favor reduction at source, but this establishes a tension between the payer’s need to have information timely and the commercial sensitivity of that information, particularly where one or more intermediaries are present.⁶³ Following work by the informal consultative group on CIV treaty claims and administration, the OECD developed the tax reporting and compliance enhancement system (TRACE), which is designed to facilitate withholding reduction at source while protecting sensitive information from disclosure to other intermediaries, largely along the lines of the US “qualified intermediary” (QI) program.⁶⁴ TRACE provides in essence that source jurisdictions may approve “authorized intermediaries” (AIs), which collect investor information and provide it to the AI country’s TA. The AI country’s TA then shares that information with the source-jurisdiction TA, which, in turn, verifies investor treaty eligibility with investor-jurisdiction TAs. The AI can then provide payers with TA-certified pooled withholding information, thus removing payer withholding risk while providing TA compliance transparency and preserving commercial confidentiality.⁶⁵ No jurisdiction has yet adopted the TRACE scheme, but subsequent technological developments (including blockchain) allow industry to manage the considerable data necessary to implement TRACE, and TAs should have an interest in TRACE if that data can be standardized in a nonproprietary format.

b. EU parent-subsidiary directive

The EU’s four freedoms may be affected by members’ rules of taxation and thus may be the subject to EU legislation. The EU PSD is designed to eliminate any disadvantage to cooperation between companies of different EU member states, as compared with cooperation between companies within the same member state, by eliminating WHT within the EU on distributions from subsidiaries of previously taxed profits.⁶⁶ Thus, the EU PSD confers a benefit similar to a tax treaty and raises similar access issues, albeit with EU-specific variations. Many private equity funds hold investments in EU PortCos via EU-resident IHCs that claim the EU PSD.

The EU PSD is not self-executing and requires local implementing law. To qualify for the EU PSD, a distribution must be paid to a company that is both EU resident and owns 10% or more of the issuer, subject to a general anti-abuse standard. “Residency” means resident of an EU Member State under the laws of that Member State, and not resident outside the EU under the terms of a double taxation agreement concluded with a third state.⁶⁷ Notably, a typical private equity fund investor does not own a 10% interest in any PortCo and so would not be able to claim the EU PSD if its proportional interest in PortCo were held directly.

The EU PSD does not preclude domestic anti-abuse legislation.⁶⁸ However, the European Court of Justice has delineated some boundaries for permissible anti-abuse approaches. For example, wholly artificial arrangements may be abusive,⁶⁹ making it important for a taxpayer to demonstrate important nonfiscal reasons for the challenged arrangement.⁷⁰ Evaluation of abuse generally must take into account the facts and circumstances,⁷¹ although ownership of an entity (e.g., by non-EU persons) generally is not relevant for this purpose,⁷² except to the extent that ownership includes a jurisdiction deemed uncooperative with the information exchange.⁷³ EU jurisdictions are free to design their tax systems in ways that do not impose WHT,⁷⁴ although non-WHT arrangements may incent misuse of the EU PSD.

The European Commission has sought to develop anti-abuse standards of general application. ATAD 1 introduces, among other things, a general anti-avoidance concept that requires EU member states to ignore arrangements that have a main purpose of obtaining a tax advantage that defeats the object or purpose of applicable tax and are not established for valid commercial reasons that reflect commercial reality, and the Danish cases arguably represent a step toward incorporating something like the OECD’s PPT into the EU PSD.⁷⁵

ATAD 3 as proposed identifies three conditions that together create a rebuttable presumption that an entity is insubstantial: (1) most of its income is “passive” income, such as dividends, interest and royalties; (2) most of its assets and transactions are cross-border; and (3) the entity outsources the administration of day-to-day operations and decision-making on significant functions.⁷⁶ Once subject to the presumption, a wide variety of additional information must be reported, including whether the entity has dedicated premises, an EU bank account, board directors meeting certain criteria and proximately located full-time equivalent employees. Failing to rebut the presumption results in disregard of the insubstantial entity and denial of a TRC.⁷⁷

Investment structuring for non-CIV funds and, in particular, the use of IHCs, has evolved with legal standards (and perceptions of legal standards) for substance, as discussed supra. When considering how private equity fund structuring may evolve, it is useful to consider the development of commercial practice regarding IHC employees, premises, and board composition and operation, three of the facts and circumstances commonly considered under substance analysis even before the proposal of ATAD 3.

As discussed above, the incompatibility of fund IHC structures with investor-level treaty claims commonly has led to the use of IHCs intended to be eligible for tax treaty or EU PSD benefits.⁷⁶ Historically, funds have held each asset in a separate IHC, which typically is resident in a jurisdiction other than the IM’s. Ordinarily, the IHC does not conduct a large amount of activity over its life, even if some of that activity is particularly important; the most consequential matters include purchase and disposal decisions, and managing its interest in the asset
(appointing and possibly instructing PortCo board members, voting its interest, pledging its interest, etc.), whereas less consequential matters that still are not extensive may include monitoring, bookkeeping, cash management and meeting such regulatory requirements as filing statutory accounts.

Less consequential IHC operational matters typically are outsourced. Dedicated office space ordinarily would not truly be needed in order to perform its functions, but practice here has varied. Some have leased dedicated office space in order to support a claim of physical presence, whereas others have viewed premises as inconsistent with the IHC’s operating model and thus as potentially problematic, rather than helpful. Board composition and operations typically remain the items that merit careful attention. Ideally, IHC boards consist of persons qualified by experience and involvement in the business to make decisions; at least half of such persons reside in the same jurisdiction as the IHC, and the board holds physical-presence meetings in the IHC’s jurisdiction for significant decisions. Even though an IHC’s board ordinarily would include at least one person from the IM’s investment team dedicated to the relevant PortCo, some members balk at traveling to board meetings, and finding qualified independent and trusted local persons for at least half of the board can prove difficult.

When there are many SPVs, the aggregate of the day-to-day functions may warrant dedicating a local person or a team to oversee the SPVs’ operations. Most commonly, this has taken the form of the IM creating an office in the SPV residence jurisdiction, staffed by personnel who manage the day-to-day operations and often serve on SPV boards. The IM may then make its office space available to the SPVs, and the staff may provide some of the administrative services needed (e.g., bookkeeping and cash management) and oversee outsourcing of the rest (e.g., preparation of accounts), all paid for by the SPVs in the aggregate under a cost-sharing arrangement. Many view this as an incremental substance improvement relative to third-party outsourcing, but the core issue remains whether, or to what extent, the IM’s premises and employees are attributable to any given SPV. This remains the case even if the SPVs are formed and meant to be resident in the same jurisdiction as an IM entity that employs investment personnel and senior management. Further improvement has been sought by itemizing costs within each SPV’s accounts (rather than a single cost-sharing charge ultimately received by the IM) and by registering each SPV as a fractional employer for legal and tax purposes (even if a single entity, such as the IM, serves as a single paymaster/payment agent).

More recent developments in legal entity structures have consolidated per-asset SPVs under a single primary IHC that employs personnel and secures office space, thus bringing these elements of substance more clearly within the IHC ownership chain. Although it is not entirely clear whether or to what extent a parent entity’s premises and employees can properly be attributed to a subsidiary, any determination that an SPV is insubstantial would, under many anti-avoidance regimes, leave the analysis to focus on the primary IHC.⁷⁸

In view of the various participants’ interests, we see four general paths for policy development concerning private equity fund structuring: (1) continue on the current path, (2) develop bright-line substance rules for IHCs, (3) develop a look-through/proportional benefits regime and (4) reduce statutory tax obligations on investment income and gains. No one option is clearly a “silver bullet” that reconciles all the goals of all participants, but each option can be evaluated for its appeal and limitations.

Continue on the current path. In the first path, continuing with current developments in various parts of the world generally would see in the EU (and possibly LatAm) continued use of IHCs seeking to claim tax treaty or directive benefits as substance standards are given more stringent content; in the United States, continued general absence of CGT for nonresidents and difficulty of treaty access due to tight LoBs; and in Asia-Pacific, varied approaches ranging from relatively low statutory tax burdens on nonresident investment (as in India and the People’s Republic of China), to bright-line substance rules (as in Singapore) and preclearance (as in Australia).

Although most well-advised private equity platforms with a global reach understand their structuring needs and the attendant tax risks, “substance” standards as currently applied do not yet have enough content to provide clear and consistent guidance, never mind guidance that takes commercial needs and practice into account. The resulting uncertainty creates additional expense for IMs (e.g., in-house expertise and ongoing management) and TAs as IMs seek to determine acceptable practices, and an uneven playing field for IMs based on their willingness to take tax risks, rather than on their ability to make and steward investments.

Develop a bright-line substance rules for IHCs. The second path would be to promulgate bright-line substance rules that identify a clear path to tax treaty and government directive qualification and are consistent with commercial practice.

Clarity increases certainty and reduces risk, which levels the playing field for IMs and reduces cost and complexity to the extent possible for all parties. However, this certainty comes at the cost of limiting regulatory flexibility to address facts that may not have been contemplated when the rules were drafted, which can lead to unintended consequences.

Develop look-through/proportional benefits regime. A third possibility would be a look-through/proportional benefits regime that enables funds to claim treaty benefits on behalf of their treaty-qualified investors without investor filing obligations.

The primary tax policy argument favoring this approach is investor equivalence, a claim that is somewhat weaker for private equity than for CIVs (see discussion supra text accompanying notes 15-16). It would also bring the same clarity and certainty associated with a rule. Notably, proportional benefits treatment most likely would largely eliminate EU PSD claims in private equity because, typically no single fund investor owns a 10% interest in PortCo via the fund.

A proportional benefits regime would need to contend with policy, legal and administrative challenges. The key policy challenges are incidence and deferral. Incidence can be addressed (and in many cases already is addressed) within the fund documents. Deferral is largely addressed by a combination of using IRR as the primary metric of investment returns and GP incentive compensation, and the typically limited life of private equity funds. Perhaps against this background, there is anecdotal evidence of TAs applying look-through treatment on an ad hoc basis in an audit settlement.⁷⁹ The primary legal difficulty with wide adoption (at least within the EU) of a proportional benefits rule appears to be the inherent likelihood of discriminatory treatment based on ownership. There is no easy response to this problem, other than that a proportional benefit rule could reduce the reliance on EU-resident IHCs. With respect to the administrative challenges, the TRACE system provides a template, and developments in regulatory information requirements (FATCA/CRS/KYC) and technology make it increasingly viable.

Reduce statutory tax obligations. The fourth possibility would entail reducing statutory tax obligations to a level around 10. Obviously, this would be a significant fiscal and political choice for source jurisdictions; while a simple solution, it may prove difficult to implement across the globe.

Against the background of the preceding discussion, we do not foresee any imminent dramatic policy changes on the horizon; instead, we expect to see incremental development of substance standards, particularly within the EU. ATAD 3 has been proposed, but it is impossible to say how it will develop. In the meantime, structuring approaches to the resulting uncertainty and risk need to focus on developing and deploying substantive entities, likely in a wider variety of jurisdictions. Increased uncertainty may warrant greater consideration of the use of transactional tax liability insurance.

Inevitably, the discussion about treaties starts from the OECD Model and the extensive reports and commentary that inform it. This appendix is designed to provide an overview of the many different reports and projects that inform the discussion in this note.

The OECD members are the world’s most developed jurisdictions. All are engaged in extensive global trade, so there is a fair expectation that all members will both import and export capital. The OECD’s Committee on Fiscal Affairs is responsible for work on all tax matters and historically a primary focus of this work has been the OECD Model Convention, including commentary and related analytical reports, all of which are designed to assist with negotiating and interpreting treaties. The OECD Model Convention has no legal weight, but it is regularly sought out as a guide to the shared understanding of common treaty provisions.

Developing countries may apply a different calculus to cross-border tax policy than most OECD economies. In 1980, under the auspices of the United Nations, non-OECD jurisdictions published a model tax convention that tends to favor the considerations of “source” jurisdictions that are much more likely to import capital than to export capital (the UN Model Tax Convention). In consequence, the UN Model Tax Convention “favours retention of greater so-called ‘source country’ taxing rights … as compared to those of the ‘residence country’ of the investor.”⁸¹  The UN Model does not have the extensive commentary and reports of the OECD Model, but the fundamental tax policy issues are similar. This note makes extensive reference to the OECD Model associated material and other OECD projects.

The most relevant OECD study points for this note concern treaty shopping, defined as using an entity to access a treaty whose benefits would not otherwise be available to the entity’s owners.^81.1 The earliest reprinted reports on this topic are those issued in 1986 concerning base companies⁸² and conduit companies. The fundamental concern with base companies was deferral — shifting income to low-tax jurisdictions, where it would remain largely untaxed, at least until repatriated, and possibly indefinitely. Very generally, the concept was to focus on transfer pricing (insisting on alignment of income with activities giving rise to the income), and anti-deferral regimes, such as CFC type rules.

The OECD issued its first report on “tax havens” in 1980, and this was followed by another in 1987, which appeared as part of a volume of four related studies.⁸³ At the direction of the G7 in 1996, the OECD opened a project on “harmful tax competition.”⁸⁴ Originally, this focused on the potential “race to the bottom” fomented by low-tax jurisdictions, but the project expanded beyond OECD membership with the development of the Forum on Harmful Tax Practices (created as a subsidiary body of OECD’s Committee on Fiscal Affairs). Following US government reconsideration under the George W. Bush administration of the propriety of condemning tax competition,⁸⁵ the project evolved to focus more on transparency and information exchange and reconstituted the wider group as the Global Forum on Taxation, which adopted standards and principles for effective exchange of tax information. The Global Forum on Taxation later became the Global Forum on Transparency and Exchange of Information for Tax Purposes.⁸⁶ Within the OECD, work continued to focus on the improper use of tax treaties, and, in 2002 it published a report on tax treaty abuse, largely driven by the harmful-tax-practices project.⁸⁷

Other OECD and OECD-adjacent efforts have focused on claims for treaty relief by CIVs. Following an OECD-sponsored industry roundtable in 2006, the OECD established an “informal consultative group” to consider both the availability of treaty benefits to CIVs and the administration of such claims. The group’s 2009 report on the availability of benefits was submitted to the OECD, which, in 2010, produced a report that resulted in additions to the OECD Model commentary.⁸⁸ The group’s 2009 report on the administration of such treaty claims also went to the OECD, which then opened a project that became the TRACE system of AIs.⁸⁹  This effort culminated with the 2012 publication of a TRACE “implementation package” designed to facilitate reduction of withholding at source (or expeditious refunds) without obligating intermediaries to disclose their investors to payers or other intermediaries.⁹⁰ TRACE is entirely about administration and has nothing to do with substantive determinations of treaty entitlement, which the group acknowledged would tend to hinge on fiscal-transparency determinations in the case of CIVs.⁹¹

The OECD’s BEPS had its germ in the 2008 great financial crisis, but OECD work began in earnest with the February 2013 report titled “Addressing Base Erosion and Profit Shifting,” and the OECD and G20 endorsed a 15-point action plan in September 2013. BEPS refers generally to tax planning strategies that exploit gaps and mismatches in tax rules, either to artificially shift profits to low- or no-tax locations where there is little or no economic activity or to erode tax bases through deductible payments, such as interest or royalties.⁹² It included several areas of study, including most relevantly for present purposes, BEPS Action 2 (“Neutralizing the Effects of Hybrid Mismatch Arrangements”), BEPS Action 5 (“Harmful Tax Practices”), BEPS Action 6 (“Prevention of Tax Treaty Abuse”) and BEPS Action 15 (“Multilateral Instrument”). The various BEPS Action Plan final reports were issued in 2015. BEPS Action 6 included work on CIV treaty eligibility and qualification for the PPT introduced to the OECD Model Treaty by BEPS Action 6, although the BEPS Action 6 Final Report acknowledged that more work was needed on non-CIV funds.⁹³ The work on non-CIV funds produced commentary and examples.⁹⁴

The various BEPS actions resulted in changes incorporated in the 2017 OECD Model. In order to streamline widespread adoption into existing treaties of the new minimum standards and other developments, without the lengthy and unwieldy process of seriatim negotiation with each individual counterpart, the OECD created a novel multilateral instrument (the MLI) that can incorporate changes across all of a jurisdiction’s treaties where counterparties do the same. An MLI “tracker” of elections made by participating jurisdictions may be found at oecd.org/tax/treaties/mli-matching-database.htm.