Featured Articles

A World Tax Court: The Solution to Tax Treaty Arbitration

Posted on September 14, 2016 by Heyka, Jake

Jake HeykaJake Heyka is a December 2017 BCL/LLB candidate at the McGill Faculty of Law, ACIArb certified, and the assistant to the co-general reporters for the International Fiscal Association Cahier on BEPS for fall 2017.

The author would like to thank professor Andreas Oestreicher of the Georg-August-Universität Göttingen for guidance in understanding the European Union's system of tax treaty arbitration, and professor Allison Christians for introducing him to the topic.

This article was selected as a winning entry in Tax Analysts' annual student writing competition.

In this article, the author examines tax treaty arbitration standards while demonstrating that as a matter of fundamental justice, arbitration should be revamped. He proposes the creation of a world tax court as a solution.

* * * * *

I. Introduction

The institution of international tax treaty arbitration (ITTA)1 is hotly debated in international business and tax law. While the process is helpful because it pressures governments to resolve contested tax decisions, opponents have called it "secret and evil."2

The merits of this last-resort, and often mandatory, instrument in double tax treaties will be debated so long as the instrument exists. However, the use of ITTA ultimately frustrates the resolution of tax disputes and should be supplanted by a world tax court.

Inquiries into the nature of ITTA are crucial to ensure the rule of law is exercised and that this relatively new process of dispute resolution develops in a way that meets taxpayer needs. Potential exists for multimillion-dollar international tax decisions to be made behind closed doors without developing any precedent and without sharing any knowledge about best practices. Often, participating parties are bound to secrecy about the process, which is not helpful for the development of good law, order, or tax governance.

Simply put, ITTA is secretive, with little public information about how to pursue an arbitration or about arbitrations that have taken place justified under the guise of protecting taxpayer information. That secrecy is unfair to taxpayers, who must already navigate a difficult set of bilateral tax treaty arrangements and domestic tax codes.

What little information does exist consists primarily of tax treaties, bilateral technical agreements, and articles about supposed arbitrations leaked to the media. Ultimately though, ITTA is a mysterious black box of extra-judicial action in need of a thorough exposé.

This article examines tax treaty arbitration standards while demonstrating that treaty arbitration should be revamped. It reviews those topics, linking the desires of ITTA founders to the present day and concluding that a world tax court is a feasible way to solve ITTA problems.

II. Arbitration Generally

Arbitration resolves disputes in a binding manner outside state-sponsored court systems. It is done most often with an odd number of arbitrators. Arbitration tribunals choose procedural rules as well as state-sponsored laws under which they have authority to make binding decisions.3 The origin of the power to be an arbitration tribunal varies, however, by the type of arbitration at play.

A. Reasons to Use Arbitration

Arbitration as a method of dispute resolution has several benefits.4 The first involves resources. Disputes in the formal court system can take lots of time and money to resolve, often for a complicated result. Most arbitrations take 10-16 months from claim to award, depending on the size of the claim.5 Arbitrators are hired for the specific dispute, tend to have expertise on the subject at hand, and can save the parties court and attorney fees.

The second advantage is privacy.6 Individuals and companies are generally able to avoid publication of any outcome because arbitration is private, and many guidelines require that arbitrators keep all information confidential.7 Individuals and corporations alike value privacy, especially when high-profile people or businesses are involved. Some information is occasionally made public, but most is never discussed outside the arbitration proceedings.8

Last, private parties tend to value the overall freedom associated with arbitration, including choice of law and jurisdiction. Given the supranational nature of international arbitration, no one type of law or jurisdiction applies unless the parties contractually agree or are bound by a treaty. Parties are also free to choose procedural rules from an arbitration institute such as the London Court of International Arbitration and the International Chamber of Commerce, to have an ad hoc tribunal in which the procedural rules are determined as the parties proceed, or to have procedural rules described in the treaty or contract.

Unfortunately, none of those benefits is always true. For example, arbitrations involving Yukos Oil that started in 2007 only recently concluded as a result of the many cross-claims and requests to set the award aside.9 Prolonged proceedings like that can lead to costs at least as high as those in a typical judicial proceeding.

B. Types of Arbitration

    1. Commercial
Commercial arbitration is used in private dealings between parties who decide through a contractual arbitration clause to use arbitration to resolve the dispute rather than the state-sponsored court system.10 Thus, the law in a commercial arbitration is first that of the contract, and then of whatever jurisdiction the parties identify as governing the contract.

Generally, one to three arbitrators decide on a resolution after supporting documentation has been presented by the parties' lawyers. The arbitrators are chosen based on various contractual criteria, or by party agreement once the dispute has begun. Most often, arbitrators have expertise in the field in which the dispute has arisen.

Typically, an arbitration panel will use rules designed by arbitration centers around the world that clear up the procedural vacuum that exists when acting outside a state-run judicial system.11 However, arbitration is highly malleable, and parties can contractually agree to the entire resolution procedure. They can also agree to have an ad hoc tribunal with the rules decided contractually or at the time of the dispute. The procedural rules will determine what submissions can be made, what evidence can be brought and how it can be presented, who can be an arbitrator, and what language the proceedings will be in.

The New York Arbitration Convention sets out the methods for requesting that a state enforce an arbitral award.12 Often, a state will adopt the U.N. Commission on International Trade Law model law or something similar to reflect the enforceability via the New York Convention.13

    2. Investment
Rather than arising out of contract between parties, international investment arbitration arises under bilateral investment treaties (BITs) between countries.14 Many countries have BITs that dictate how a foreign investor from State A can be protected to guarantee that State B does not expropriate the investor from A and to ensure the investor receives fair and equitable treatment. Like tax treaties, BITs are agreed upon through lengthy negotiations to set the standards for the interactions between the jurisdictions and its citizens.15

Most often, the panel includes one or three arbitrators who have expertise in the matter.16 The most important item regarding arbitrators, at least from the perspective of the International Centre for Settlement of Investment Dispute (ICSID), is certainty that most of the arbitrators are not nationals of the jurisdiction party to the dispute or the state the investor is from.

As in commercial arbitration, international procedural rules have been established for assisting the state and the investor in resolving the dispute. Unlike commercial arbitration, however, the procedural rules are chosen in the BIT rather than a contract, so any investor investing from A into B consents to the use of the procedural rules in the BIT simply by investing in B.17 The BIT sets out the terms for enforcement and may be supplemented by domestic legislation.

    3. State-to-State
A less common version of arbitration is that between states, a modified version of diplomacy whereby a citizen asks his home country to resolve a legal problem in another country on his behalf. There are no procedural rules because the diplomatic mission of one county will often work to resolve the dispute with representatives from the foreign state in a simple conversation. In many ways, it is more like mediation; however, its private nature and the agreements resulting from the diplomatic discussions put it under the umbrella of arbitration rather than mediation.

III. History of ITTA

Tax arbitration is an odd concoction of the above methods of arbitration.18 It developed through a confused history that resulted in the four major ITTA systems around today: the OECD base erosion and profit-shifting models, the U.N. model, U.S. models, and the EU arbitration convention.

The idea of tax arbitration traces back to the 1970s, when Gustaf Lindencrona and Nils Mattson proposed the creation of a tax arbitration court following the creation of the OECD Model Tax Convention on Income and on Capital, a keystone in the foundation for modern tax treaties.19 Countries had long argued over how to tax individuals engaging in cross-border commerce, and those treaties sought to settle those disputes.20 Mutual agreement procedures (MAPs) were one of the primary methods for resolving disputes, generally found in article 25 of tax treaties. Unfortunately, MAP is nonbinding and truly functions only as mediation, thus yielding the possibility that a dispute does not settle, leading to undesired consequences for the taxpayer and leaving serious law and revenue questions unanswered.21 Without consistent tax decisions between countries, the purpose of the tax treaty is defeated.

After MAP failed, competent authorities struggled to use versions of arbitration without a central authority or real guidelines regarding how to resolve a tax dispute. Lindencrona and Mattson then suggested a world tax court with the United Nations as the arbiter. They believed a supranational body like that sort was the only way to ensure fair and just taxpayer treatment. They realized, however, that such a radical move was not a reasonable prospect at the time. It was simply unfeasible to create a world tax court under the auspices of any international organization, given the political climate surrounding the use of international law. Accordingly, they settled on the idea of arbitration as a stepping stone to the eventual creation of a world tax court.

The idea was not implausible. Major players in the world of international tax policy development such as the International Fiscal Association and the International Bar Association had already expressed interest in arbitration-like procedures to resolve tax disputes.22 Knowing that, Lindencrona and Mattson asked international tax observers to encourage the use of arbitration clauses in tax treaties, which went unheard for quite some time.

In 1985 the International Chamber of Commerce (ICC) indicated its interest in the creation of ITTA, although its statement received little response. The same year, Germany and Sweden negotiated a tax treaty that included an ITTA clause.

The EU first formally adopted ITTA in the form of a multilateral agreement between a few member states in 1990.23 The EU arbitration convention (90/436/EEC) did not come into force until 1995 (but was not used until later, for reasons that are unclear) and was initially set to end in 2000.24

In 1999, in the middle of the EU's stall on implementing the EU arbitration convention, the ICC revisited its policy statement from the 1980s and stated its commitment to "encourage governments to accept compulsory arbitration in international tax conflicts."25 In 2002 the ICC released another report praising tax arbitration, this time presenting a draft article countries could use in their tax treaties to permit ITTA.26 As with its 1980s policy statement, the ICC's thoughts played little role in the development of ITTA.

In 2004 the EU Joint Transfer Pricing Forum finally revisited the idea,27 and the EU and 15 member states ratified the treaty with retroactive effect to 2000.28 Since then, most EU member states have agreed to follow the EU arbitration convention if MAP has failed, although many have reservations about its use.

ITTA didn't really hit the global stage until 2007, when the OECD developed its model for tax arbitration agreements, and in 2010, when the OECD released its modified model tax convention.29 Through the new model treaties, countries outside the EU and EU countries that dealt with states outside the EU had standardized guidance on how to use and implement an ITTA clause. In 2007 the United States began to add the OECD's article 25(5) to its tax treaties.30 A few other countries also responded but the United States has mostly led the charge on creating ITTA provisions outside the EU. Many countries -- apart from those with which the United States has memoranda of understanding on tax arbitration -- have not shown that their arbitration clauses are functioning.31 Many others' treaties are signed but not yet in force.32

The U.N. model on tax treaties did not include an ITTA provision.33 However, in 2011 the U.N. adopted its version of the arbitration procedure and guidelines as part of its model tax treaty.34

The latest in the ITTA saga is the OECD BEPS project, whose 15 action items identify changes that need to be made to stop BEPS. Action 14 suggests broad implementation of MAP and arbitration clauses in tax treaties. Around 15 countries have committed to putting ITTA clauses in their tax treaties, a major step in normalizing ITTA.

IV. Modern ITTA

Modern ITTA exists as a consequence of failed MAP. While it is akin to other types of arbitration, it does not adhere to any one standard and has become a niche and poorly understood dispute resolution regime. At its core, it is a version of state-state arbitration but has also adopted characteristics of other methods. In general, the competent authorities of two countries convene a panel of arbitrators that renders a decision -- much like standard commercial arbitration, except the parties are the competent authorities acting on the taxpayers' behalf. Taxpayer involvement varies, as does the amount of information the taxpayer is privy to at the end of the arbitration.


In 2007 the OECD proposed adding a fifth paragraph to article 25 of the OECD model convention to provide that in the event of a treaty dispute and MAP failure, a taxpayer could request arbitration, set up by the competent authorities of the countries concerned. While the OECD model states that ITTA is not meant to supplement MAP, it does, and its model has been extensively discussed.35

For purposes of this article, the following points are important:

  • If MAP fails, a taxpayer can request ITTA. One must endure two years of stalemate in the MAP process before requesting ITTA. After all documents have been submitted, ITTA ideally takes no longer than six months.
  • The governments of the two states engaged in ITTA bear the costs.
  • Arbitrators are qualified based on affiliation with the competent authorities of the states engaged in the ITTA but have no other requirements. Each competent authority will appoint an arbitrator, and those arbitrators jointly appoint a third arbitrator to be panel chair.
  • The competent authorities present their case to the arbitrators, who render a decision. Decisions are not intended to be made public but can be if the taxpayer's identity is kept private and it is clear the results do not have strict precedential value.
  • The taxpayer can make a written submission, have representatives, and make an oral pleading if the arbitrators agree.
  • Procedural rules are determined ad hoc.
  • The model does not address enforcement except to note that state parties are bound by the decision.
  • Generally, the taxpayer can choose whether to accept the decision after it has been rendered under the concept of the "last best offer." Rendering a simple decision is also permitted and binds the taxpayer without a right of refusal.36

The 2015 BEPS action 14 report makes no major changes to the OECD 2007 model. Its primary contribution is stating that countries should adopt both MAP and ITTA procedures. It points out that ITTA transparency is a problem and states that suitable alternative provisions will be developed. On a positive note, action 14 purports to set up a platform for localizing all information about ITTA provisions to build a body of information on the topic.

B. U.N. Model

The U.N. model differs from the OECD model in a few ways.37 First, it provides for two model clauses: one with only MAP and one with both MAP and arbitration. The hold-out period before one can ask for arbitration is three years instead of two. The taxpayer is even less involved in the process because the onus is on the competent authorities to request arbitration. Last, the competent authorities are not bound by the final decision if they make a side agreement, so long as that is done within six months of the arbitral panel's decision.

C. United States

U.S. versions generally follow the OECD 2007 model. The United States most often references ICSID fee schedules for the remuneration of arbitrators. Decisions are never made public.38 Further, the U.S. models adopt the last best offer method, which allows the taxpayer to choose the decision.39 Last, instead of providing a reasoned opinion, the United States mandates that the decision will be stated without any explanation of how or why it was rendered.40

D. EU Model

The EU arbitration convention was created long before the OECD model but generally follows it. It diverges in the following ways:
  • only countries that are signatories to the EU arbitration convention may use its procedures, and only then between states that have acceded to the convention;41
  • an EU panels consists of five or more arbiters chosen from a public list of "Persons of Independent Standing" and all of whom are recognized as tax experts in their countries;42
  • the EU arbitration convention only addresses transfer pricing assessments;43
  • the EU arbitration convention gives a three-year limit after which the taxpayer cannot request arbitration;44
  • EU arbitrations are not supposed to take longer than three years;45 and
  • the decision, once rendered, is automatically applied to adjust the prior tax decision.46
V. Problems With ITTA

Questions arise both when looking at ITTA from normative and historical perspectives, and when examining it with an understanding of the justifications for arbitration as a dispute resolution mechanism.47

A. Confidentiality

Arbitration both in and outside the tax context is private. Little documentation is released, and the public rarely knows about arbitrations. Thus, little is known about ITTAs.

What we do know lies in a few key documents, such as treaties and supporting procedural guidelines on how to make a submission, that guide a taxpayer into an ITTA.48 Some countries -- perhaps atypically -- provide the documentation that explains ITTA methods, but only to the degree that their bilateral agreements allow. That practice is problematic because the law is meant to build on itself, which it cannot do if everything is kept secret. Further, it negates the possibility that a taxpayer could have any guidance on how to approach the arbitration apart from the few treaty documents that exist.

Secrecy has its advantages in commercial situations but is problematic when looking to develop common law for the resolution of tax disputes. In the United States, for example, constant regulations and clarifications are released about how to apply various tax laws. Why not do the same for arbitration? It makes sense to keep the private dealings of two parties secret in commercial arbitration because they are the only parties concerned with the dispute. But when venturing into the realm of public law where people rival the state, respect for the concept of stare decisis becomes key so that citizens can have guidance regarding how the law is likely to affect them in their dealings with the state. ITTA privacy concerns could be solved by redacting party names as well as dollar amounts at issue.49

Justifying privacy under the guise that there is no comparability between ITTA cases as there is in commercial arbitration is without virtue. Many investment arbitrations, the closest cousin to ITTA, are made public to show the process and to provide information regarding how future cases under BITs might be decided.50 Good governance requires that taxpayers have guidance on how they can fight a state when they disagree with the way it has treated them, or face a potential six-year waiting period before they know whether they are liable for taxes.51 Further, BEPS action 14 synchronizes arbitration clauses in tax treaties, and action 15 synchronizes tax treaties as a whole, making the comparability of cases even more plausible and the benefits of ITTA public disclosure more widespread.52

Moreover, when proceedings are secret, taxpayers have no check to ensure that ITTA tribunals are treating them fairly. Nor can they ensure that the process is not massively corrupt, with payoffs to receive a favorable result. Without any public knowledge of the proceedings, taxpayers have no way to criticize the process and demand its improvement.

The only available information about ongoing ITTAs is an occasional MAP statistic.53 However, only the EU statistics include information on whether arbitrations are happening at all, leaving taxpayers in the dark on yet another front regarding ITTA.54

B. Taxpayer Submissions

That a taxpayer and his lawyer are not allowed to play a role as advocates is yet another problem with ITTA. In many commercial and investment arbitrations, the claimant and his lawyer are involved from the claim to the award, with full rights to submit evidence, briefs, counterclaims, and other documents. ITTA does not follow standard arbitration procedures, allowing a lawyer and taxpayer only to submit a request for arbitration to a competent authority. In most ITTAs, a taxpayer can submit more than the initial claim only on request from the arbitral tribunal. Cutting the taxpayer out of the process heightens secrecy and negates the taxpayer's ability to advocate for himself.

C. Fees

Many arbitration guidelines from tax authorities cite an institutional fee schedule to determine arbitrator compensation in ITTA, although action 14 does not mention fees.55 Those pay scales are confusing because the taxpayer has no obligation to pay the governments to resolve the dispute; public tax dollars pay arbitrator fees.56 Further, it is curious that government officials from competent authorities are paid the same way and rate as a private arbitrator in the investment context. That could provide an incentive for competent authority representatives to drag out the process and earn more money.

D. Arbitrator Impartiality

A fact-finder's impartiality is always a concern, and is exacerbated when most of the arbitrators are government officials whose daily goal is to reap revenue from taxpayers -- in nontax arbitration, using party-appointed arbitrators is standard. A competent authority's incentive is not to resolve the tax dispute but to get revenue from the taxpayer by whatever means possible. Decisions are made by majority, so the competent authority representatives could overpower the state arbitrator's vote by side agreement, or cause difficulties in the process by blatantly representing state interests rather than considering the taxpayer's interests -- as has been happening in the EU.57 The taxpayer might reject a decision that results in higher taxation, but because it first turned to arbitration after finding the original tax decision unjust, that is not necessarily a reasonable resolution.

E. Length, Appeals, and Enforcement

The potential length of ITTAs is troublesome. There are no statistics on how long ITTAs take, but there is anecdotal evidence that it can be more than three years.58 Prolonged uncertainty about one's tax consequences is inefficient and runs counter to a core benefit of arbitration: a process quicker than that of a normal court system.

Further, if after several years an arbitration panel hands down a decision that is innately incorrect, the taxpayer has no appeal process because the whole endeavor must be kept private. Thus, the taxpayer must choose between the lesser of two evils: being subject to double taxation or being unable to speak of the issue again.59

Enforcement was another major concern in arbitration before passage of the New York Convention. Now, one need only appeal to a national court for an arbitral award to be recognized and enforced. With ITTA, tax treaties will state in article 25 that the taxes will be adjusted, but if the process is secret, how can a taxpayer force anyone to make corrections?60 Outside the EU, the answer is unclear.

F. Developing Countries

The above questions are even more problematic for developing countries.61 If countries with the resources to counter those problems are struggling to do so, developing countries hoping to find a way to resolve international tax disputes will have an even more difficult time.

VI. The Solution: A World Tax Court

Standardizing ITTA will create some procedural certainty but does not guarantee consistent use of those procedures, allow the public to see whether the process is fair, or establish reliable precedent. As Lindencrona and Mattson suggested over 30 years ago, ITTA should be a stepping stone to what the world ultimately needs: a world tax court.

As radical as it may seem, the idea is not far-fetched. World courts exist in many commercial and noncommercial contexts, and those that deal with money rather than crime are followed by many countries and used quite often.62 Moreover, state authority is regularly ceded to resolve disputes between commercial parties in arbitration courts such as the Permanent Court of Arbitration in The Hague, the London Court of International Arbitration, and many other arbitration institutes. A world tax court would merely serve as a place to resolve tax disputes in a similar manner while sustaining the public nature of tax law.

A. Hosting the Court

The OECD is the natural candidate to host the world tax court. It has served as a global fiscal policy research and development body since the 1960s, and its Centre for Tax Policy and Administration leads the charge on standardizing international tax policies worldwide.63 It does not seem that the OECD would shy away from this opportunity, given the recent strides it has made to fortify its role in international tax.

The world tax court proposal becomes even more feasible when examining the BEPS project beyond the scope of having counties adopt ITTA via the action 14 model. Action 15 is dedicated to developing a multilateral tax instrument to standardize tax treaties, making the domestic tax policy of the countries involved the only difference among disputes. That standardization would likely include elements of action 14. A court to manage that standardized body of law should be an easy sell if countries have already ceded to the OECD their authority to resolve tax disputes and agreed to a multilateral tax treaty. Accordingly, the only major step would be to set up shop and start resolving tax disputes.

Some observers -- fearing that the OECD, led by the G-20, would not recognize the interests of developing nations -- might think the U.N. Committee of Experts on International Cooperation in Tax Matters should host the court. However, the committee is small, underresourced, and far behind the OECD in its development, and its model tax treaty is directed only at relationships between developed and developing nations.

Some groups might also worry that powerful countries would try to block the creation of a world tax court. However, the United States has been behind the recent push for ITTA and has the most ITTA clauses in force outside the EU. Countries such as France are also unlikely to protest because they have agreed to the EU arbitration convention and to using the OECD action 14 model for ITTA. It has been hinted that some South American countries might protest the institution of widespread ITTA. If Argentina or Brazil -- the area's largest economies -- agree to a world tax court, that concern would be mitigated because it would encourage other countries in the region to join.64 Last, given that the OECD is a way for the G-20 to develop policy, the powerful states that might normally block a world court would have an active role in policy development, thus decreasing the likelihood that they would prevent the success of a world tax court.

D. Procedures and Benefits

Like other global courts, a world tax court would first need an establishing treaty to identify its procedures and member states. Its seat would be at OECD headquarters in Paris.

Judges or arbitrators would be like those used in ITTA, but all names would be published, not just those on the list of EU Persons of Independent Standing. That would solve the uncertainty surrounding those involved in arbitrations and promote arbiter impartiality. The panel would comprise three to five arbitrators: at least one from each state involved and one from a third country. Decisions would be by simple majority. Arbitrators would have international tax expertise and be appointed by their countries.

Arbitrators should not merely be representatives of, or chosen by, the competent authorities of each state to avoid the difficulties the EU has faced.65 Instead, current and former legal and accounting practitioners with international tax experience but no significant connection to the ministry of finance from their respective countries should be appointed.

The court should redact taxpayer information and dollar amounts from all decisions, akin to IRS private letter rulings.66 That would preserve privacy while providing the world with a developed body of law showing how a tax dispute might generally be resolved.

Default rules for procedure would be agreed on in the establishing treaty, standardized based on the multilateral instrument in BEPS action 15 and the model ITTA clause and procedures in action 14.

Further, the world tax court should have authority to run MAPs so that an entire international tax dispute could be addressed in a single location. The OECD is well versed in that area, having already developed a standard for MAP.67

Last, all signatories would cede authority to court representatives to render final decisions in tax disputes, giving the taxpayer the option to accept the decision or instead accept the prior, protested decision. As with commercial and investment arbitration, appeal could be made only on procedural grounds, such as gross misconduct by the judges.

VII. Conclusion

Ultimately, ITTA is a mixed bag of processes through which a taxpayer can ask for a review of his cross-border tax assessment that is not reaching its full potential. It suffers from too much secrecy and a mangled set of procedures. Thankfully for taxpayers, many interests are converging toward a possible solution: a world tax court. Without a court like that, taxpayer rights might not be properly represented.


1 Tax treaty arbitration, generally, is an alternate dispute mechanism for contested tax decisions in cases of international taxation of a person or entity. It is most often provided for in the bilateral tax treaty between State A and State B concerned with the given tax decision. This process, according to the standard set forth in section 25 of the 2010 OECD model tax convention, is a last resort after the mutual agreement procedure (MAP) has failed. It is not, as some often mistake, the arbitration of a tax issue in a commercial contract or under a bilateral investment treaty, though there are stark similarities to those forms of arbitration and that of tax treaty arbitration, as will be discussed in this article. For a discussion of MAP, see Allison Christians, "How Nations Share," 87 Indiana L.J. 1407 (2012). For a discussion of tax in various arbitration settings, see Alexis Foucard and Léa Grandfond, "Arbitration of International Tax Disputes: A Move Towards Democratization?" in Contemporary Issues in International Arbitration and Mediation: The Fordham Papers 2014, 419 (2015).

2 Lee A. Sheppard, June 19, 2014, presentation at the McGill Tax Justice and Human Rights Research and Collaboration Symposium.

3See, e.g., Canada's Ontario Arbitration Act, SO 1991 c 17; the U.S. Federal Arbitration Act (1925); and the U.K. Arbitration Act 1996 (UK) c 23.

4 For a discussion on the motivations behind arbitration, see Emilia Onyema, Introduction to International Commercial Arbitration 12-19 (2008).

5See, e.g., London Court of International Arbitration, "LCIA Releases Costs and Duration Data" (Nov. 3, 2015); International Chamber of Commerce (ICC), "2014 ICC Dispute Resolution Statistics," ICC Disp. Res. Bull. (2014); and American Arbitration Association, "AAA Arbitration Roadmap," at 11 (2007).

6 ICC, "Arbitration Rules: Mediation Rules," at article 21 (2013); and United Nations Commission on International Trade Law (UNCITRAL), "UNCITRAL Arbitration Rules," at article 25(4) (revised 2010).

7See, e.g., ICC arbitration rules, supra note 6, at 45-46; and JAMS, "Comprehensive Arbitration Rules & Procedures," at 28 (July 1, 2014).

8See, e.g., International Centre for Settlement of Investment Disputes (ICSID), "Cases" (2016); see also UNCITRAL, "Case Law on UNCITRAL Texts" (2016).

9 The Yukos Oil arbitration is the largest in history. For more on this arbitration, see Dmytro Galagan, "The Challenge of the Yukos Award: An Award Written by Someone Else -- a Violation of the Tribunal's Mandate?" Kluwer Arbitration Blog (Feb. 27, 2015); see also Martin Van Tartwijk, "Dutch Court Quashes $50 Billion Award to Former Yukos Owners," The Wall Street Journal (Apr. 20, 2016).

10 For a good overview of commercial arbitration, see Onyema, supra note 4; see also Christopher Drahozai, "International Arbitration Law in the United States," in International Arbitration Law: A Comparative Study (2007).

11See ICC rules, supra note 7; and London Court of International Arbitration rules (Oct. 1, 2014).

12See Convention on the Recognition and Enforcement of Foreign Arbitral Awards (1959).

13 The UNCITRAL Model Law on International Commercial Arbitration is model legislation created by UNCITRAL to help standardize the way domestic legal systems handle commercial arbitration. It sets out guidelines for elements of procedure and enforcement of awards and is often adopted in full.

14 For a good overview, see Rudolf Dolzer and Christoph Schreuer, Principles of International Investment Law (2012); Tarcisio Gazzini and Eric De Brabandere, International Investment Law: The Sources of Rights and Obligations (2012); and Scott Shackelford, "Investment Treaty Arbitration and Public Law, by Gus Van Harten," 92(1) Stanford J. Int'l L. (2009).

15 To view the BITs, see United Nations Conference on Trade and Development, "International Investment Agreements Navigator" (updated 2013). See also ICSID, "Database of Bilateral Investment Treaties" (2016).

16 ICSID, "Convention, Regulations and Rules," at Chapter I (Apr. 2006).

17 The lack of a BIT can be a major hurdle for investors if they experience one of the problems BITs normally protect against. For example, that was a problem in the Yukos Oil arbitration because the investors were from the United States and Russia had taken the company; those countries do not have a BIT in force.

18 The best comprehensive history of ITTAs is Jean-Phillippe Chetcuti, "Arbitration in International Tax Dispute Resolution," Inter-Lawyer (2001).

19See generally Gustaf Lindencrona and Nils Mattson, Summary of the Feasibility of a World Tax Court (1979). A few countries were outliers, having adopted ITTA in treaties as far back as 1959. See International Fiscal Association, "Resolution of Tax Treaty Conflicts by Arbitration," at 89 (1994).

20 The original disputes can be traced back to the 19th century; however, it was not until the 20th century that nonbinding resolution procedures were analyzed. See Lindencrona and Mattson (1981), at 13.

21 Lindencrona and Mattson (1979), supra note 19, at 23; and Lindencrona and Mattson (1981), supra note 20, at 61.

22 The IFA showed interest in 1951 and 1960. The International Bar Association first expressed interest in 1956. See Lindencrona and Mattson (1981), supra note 20, at 29-30, 51.

23See EU Convention on the Elimination of Double Taxation in Connection With the Adjustment of Profits of Associated Enterprises -- Final Act -- Joint Declarations -- Unilateral Declarations, 90/436/EEC (Aug. 20, 1990) (EU arbitration convention).

24 Acts Adopted Pursuant to Title VI of the Treaty on European Union, 1999/C 202/01 (July 16, 1999).

25See ICC, "Arbitration in International Tax Matters" (May 3, 2000), and "Arbitration in International Tax Matters" (Aug. 29, 2001).

26 ICC, "Arbitration in International Tax Matters: Bilateral Convention Article" (Feb. 6, 2002).

27 Michelle Markham, "The Resolution of Transfer Pricing Disputes Through Arbitration," 33(2) Intertax 69 (2012).

28 EU Joint Transfer Pricing Forum, "Report on the Re-Entry Into Force of the Arbitration Convention," JTPF/019/REV5/2004/EN (May 30, 2005).

29 During a January 2007 meeting, the OECD proposed adding an arbitration article to the model treaty. That meeting is recognized as the point at which ITTA truly started to gain ground. See OECD Centre for Tax Policy and Administration, "Improving the Resolution of Tax Treaty Disputes," at 12 (Feb. 2007).

30See, e.g., Canada-U.S. tax treaty (1985); and Belgium-U.S. tax treaty (2009).

31See, e.g., Ireland-Israel tax treaty (1995) at article 25(5); see Japan-U.S. tax treaty (2013).

32See Annex 1 for a comprehensive look at the countries in tax treaties with G-20 states and what MAP and arbitration clauses exist.

33 Most governments do not use arbitration because it is costly to the governments, secretive, one-sided in that only the governments take part, and an intrusion on national sovereignty; see R. Biçer, "International -- The Effectiveness of Mutual Agreement Procedures as a Means for Settling International Transfer Pricing Disputes," 21(2) Int'l Transfer Pricing J. (2014).

34 The differences are detailed in Section IV. See U.N., Model Double Taxation Convention Between Developed and Developing Countries (2011).

35See, e.g., Marcus Desax and Marc Veit, "Arbitration of Tax Treaty Disputes: The OECD Proposal," 23(3) Arbitration International (2007); H.M. Pit, "Arbitration Under the OECD Model Convention: Follow-Up Under Double Tax Conventions: An Evaluation," 42 Intertax (2014); Brian J. Arnold, "The Scope of Arbitration Under Tax Treaties," in International Arbitration in Tax Matters (2015); and Jasmin Kollmann et al., "Arbitration in International Tax Matters," Tax Notes Int'l, Mar. 30, 2015, p. 1189.

36 OECD 2007 meeting, supra note 29.

37 The U.N. model is discussed and summarized in Hugh Ault, "Dispute Resolution: The Mutual Agreement Procedure," U.N. Papers on Selected Topics in Administration of Tax Treaties (May 2013), at 21-26.

38See IRS, "Taxpayer Consent to MAP Arbitration and Nondisclosure Agreement."

39See, e.g., IRS, Memorandum of Understanding Between the Competent Authorities of the Kingdom of Belgium and the United States of America (May 6, 2009).

40See, e.g., id.

41See EU, "Revised Code of Conduct for the Effective Implementation of the Convention on the Elimination of Double Taxation in Connection With the Adjustment of Profits of Associated Enterprises," 2009/C 322/01, at article 2 (Dec. 30, 2009).

42See id. at article 7; and EU arbitration convention, supra note 23, at article 9.

43 EU revised code, supra note 41, at article 1.

44Id. at article 4.

45 This is not always the case, as indicated by the EU Joint Transfer Pricing Forum, "Final Report on Improving the Functioning of the Arbitration Convention," JTPF/002/2015/EN (Mar. 12, 2015), at 8.

46See EU revised code, supra note 41, at article 6.4(e).

47 This section concentrates primarily on U.S. ITTA agreements and the EU arbitration convention. It does not criticize the OECD or U.N. models because they serve as reference points and are ever-changing.

48 In the United States, those documents are available on the IRS webpage, "Mandatory Tax Treaty Arbitration." EU documents are less available, but most can be found on the Joint Transfer Pricing Forum site.

49 Some have claimed that confidentiality is essential, although without justification. See William Park, "Income Tax Treaty Arbitration," 10 George Mason L. Rev. 818-819 (2001). I argue that openness is essential, and Lindencrona and Mattson would agree. See Lindencrona and Mattson (1981), supra note 20, at 19.

50See, e.g., ICSID, "Cases" (2016).

51 That is a significant concern for corporations and high-net-worth individuals with millions of dollars on the line but almost no guidance regarding how an ITTA might proceed.

52 While the multilateral instrument still needs to be finalized, the OECD is working to do so and find agreement among countries. The OECD hopes to deliver the final version by the end of 2016.

53 For the latest statistics, see IRS Large Business and International Division Competent Authority Statistics (Apr. 16, 2015); OECD, "Mutual Agreement Procedure Statistics for 2014" and EU Joint Transfer Pricing Forum, "Statistics on Pending Mutual Agreement Procedures (MAPs) Under the Arbitration Convention at the End of 2013," JTPF/008/2014/EN (Oct. 2014).


55See, e.g., Canada-U.S. tax treaty, "Arbitration Board Operating Guidelines," at 10(a); and Belgium-U.S. tax treaty, "Arbitration Board Operating Guidelines," at 11(a).

56 Those fees can reach into the millions of dollars. See Michael Lennard, "International Tax Arbitration and Developing Countries," in Contemporary Issues in International Arbitration and Mediation: The Fordham Papers 2014 444-446 (2015).

57 EU Joint Transfer Pricing Forum final report, supra note 45, at 14-15.

58Id. at 10, 13 (discussing the two- and three-year periods as only a part of the entire EU arbitration convention sequence of events when a taxpayer wants to have his case heard).

59See, e.g., IRS taxpayer consent, supra note 38. The EU does not go to the same lengths to bind taxpayers to confidentiality.

60See, e.g., Belgium-U.S. tax treaty (as amended Nov. 27, 2006), at article 25(4).

61 For further discussion, see Lennard, supra note 56.

62 The International Criminal Court has had only 39 indictments and two convictions since its founding in 1998. Part of its problem is dependence on the jurisdictions that are members of its convention to bring criminals to The Hague and enforce decisions; the United States and several other powerful countries refuse to be signatories, rendering the court both toothless and spineless. See Rome Statute of the International Criminal Court, at articles 58-59, Part X (1998); and International Criminal Court, "Report of the International Criminal Court," A/70/350 28 (Aug. 2015).

The International Court of Justice and the Permanent Court of Arbitration have resolved many cases and are far closer to the model this article suggests for the world tax court. Also, ICSID had registered 549 cases by the end of 2015. See, e.g., International Court of Justice in The Hague, "Yearbook 2012-2013"; Permanent Court of Arbitration, "Annual Report 2015"; and ICSID, "The ICSID Caseload -- Statistics," at 5 (2015).

63See Jeffrey Owens and Sebastian Beer, "The Structures and Mandates of Eight International and Regional Organizations That Work on Tax," Vienna University of Economics and Business, at 39-47 (Jan. 2014); and Diane M. Ring, "Who Is Making International Tax Policy? International Organizations as Power Players in a High Stakes World," 33(3) Fordham Int'l L.J. 649 (2010).

64See Natalia Quiñones Cruz, "International Tax Arbitration and the Sovereignty Objection: The South American Perspective," Tax Notes Int'l, Aug. 11, 2008, p. 533.

65 EU Joint Transfer Pricing Forum final report, supra note 45, at 14-15.

66 The OECD also hinted at doing that in the commentary on its 2007 model, showing a willingness to publish tax rulings.

67See OECD, "Manual on Effective Mutual Agreement Procedures" (Feb. 2007).