Improving International Tax Dispute Settlement
What are international tax disputes?
Formally, international tax disputes are between the tax authorities of two different countries. They result from differing interpretations of the provisions of a tax treaty between the two countries. However, they mainly affect taxpayers with cross-border economic activities, usually transnational corporations (TNCs). Tax treaties are normally incorporated into domestic law. So, taxpayers can go to court if they disagree with how a treaty rule is applied. But tax treaties also give them the right to complain to the competent authority in the relevant national tax administration. The competent authority is obliged either to resolve the issue, or, under the mutual agreement procedure (MAP), to consult with the competent authority of the treaty partner.
Under the MAP, the competent authorities must ‘endeavor’ to solve the problem – but are not obliged to do so. The MAP is totally secret. Even the existence of a claim is not made public. Tax advisers prefer the MAP over court cases, which generally are public. But they complain that the MAP takes too long, and does not guarantee an outcome. They have long urged that unresolved disputes should go to binding third party arbitration.
The growth of disputes
Since 2006, the number of new MAP cases has doubled, and the number outstanding has more than doubled. Cases are taking longer to resolve. It seems that most cases concern the allocation of profits of TNCs among different taxing jurisdictions. A treaty between EU states has since 1995 required transfer pricing cases which remain unresolved after two years to be referred to a Commission of experts and representatives of each tax authority. The Commission produces a reasoned opinion. This can only be published with the agreement of the parties to the dispute. No such opinion has ever been published, and in fact less than half a dozen cases have ever been referred to a Commission. The main effect has been to pressurise competent authorities to resolve cases within the two-year deadline.
Ten years ago, the US introduced a different form of arbitration, for cases with Canada and Germany, also for disputes unresolved within two years. Under this ‘baseball’ or ‘short-form’ procedure, the arbitrators cannot give their own independent opinion, but must choose between the last best offers tabled by the parties in dispute. The decisions must not be published, or even cited in later cases. Around ten arbitrations have taken place.
Neither of these procedures has slowed the growth of cases or the time taken to resolve them. Arbitration has been included in the OECD model convention (2007) and as an option in the UN model (2011). Over 200 actual treaties, including some with developing countries, now have some version of arbitration. A key issue is who can trigger arbitration. Under four of the developing country treaties the taxpayer can compel arbitration if the competent authorities cannot agree, as in the OECD model. The other treaties require the consent of either one or both competent authorities. However, developing countries have had few MAP cases, and most have rejected compulsory arbitration. Their experience with international investment arbitrations, some of which have involved tax matters, has been discouraging.
Causes and solutions
Most disputes concern the allocation of profits of TNCs, and the growth has occurred as enforcement of transfer pricing rules has strengthened. Developing countries have been introducing transfer pricing rules and are now improving enforcement. So TNCs fear there will be more disputes. This fear may be justified, since the OECD Transfer Pricing Guidelines are complex and do not provide clarity. They require tax officials to identify the functions performed by each part of the TNC, by analysing its business model. This requires specialised knowledge and involves discretionary and subjective judgements.
India saw a rapid rise in disputes after it introduced transfer pricing regulations based on the OECD Guidelines in 2001. By 2012 had a backlog of over 3,000 tax tribunal cases. India does not publish MAP data, but a conflict between the US and Indian competent authorities became public, leading to the replacement of the Indian official in 2013. In January 2015, the two competent authorities signed a Framework Agreement intended to facilitate resolution of some 200 cases. A year later a statement said that about half had been resolved. In contrast, Brazil, which uses fixed transfer pricing margins, easy to administer but considered unorthodox by the OECD, has experienced little litigation and few MAP claims.
The G20/OECD project on base erosion and profit shifting (BEPS) recommended that national tax officials dealing with the MAP should be separate from frontline tax audit staff and give autonomous decisions. This is difficult for tax authorities that are already under-resourced, especially in developing countries. The proposals are designed to strengthen the international community of specialists in the MAP. It includes the private sector tax advisers, many of whom are former public officials, and is dominated by experts from OECD countries. Arbitrators are drawn from this community, and would reject interpretations that it considers unorthodox. Monitoring through peer review would reinforce this culture, and secrecy would be retained, despite public suspicions about cosy deals.
A better way forward
The aim should be to minimise conflicts by making the rules clearer and easier to apply. The MAP should be used to agree general interpretations that can be published, dealing not only with taxpayer claims of double taxation, but also issues of double non-taxation, which are especially important in this period of change. The Framework Agreement between India and the US was successful, but could have had a greater impact if it had been published.
Further, publication of the outcomes of actual MAP cases would provide guidance for other taxpayers, help ensure that like cases are treated alike, and reassure a wider public that decisions are fair. Reforms should aim to allocate the profits of TNC groups based on clear and quantifiable factors that reflect the actual economic activities and value created in each country. This would both reduce the number of conflicts and ensure that the disputes that do occur can be adjudicated in a transparent, fair and consistent manner.