As the world is increasingly interconnected, international taxation – traditionally more of a niche issue for tax lawyers – is receiving more and more attention in wider discussions on economic development: Double tax treaties, or agreements that two countries sign with one another to prevent multinational corporations or individuals from being taxed twice, have become more common, with more than 3,000 in effect today. And while they may contribute to investment, some have also become an instrument for aggressive tax planning.
Historically, countries have often neglected subjecting tax treaty policy to proper analysis of cost and benefits. However, in light of a number of aggregate studies, associated costs may be substantial. The trickiness of this issue can be illustrated with the case of Ukraine, which is elaborated on in our recent working paper, “The direct and indirect costs of tax treaty policy: evidence from Ukraine.” We hope that this initial study of Ukraine can be a first step towards providing analytics and advice to help assess costs and benefits associated with double tax treaties more systematically.
Presumed tax treaty benefits typically include increases in trade and investment flows, increased investment certainty (though, as Martin Hearson noted recently, unbalanced treaties may actually lead to more uncertainty), tools to combat tax evasion, and strengthening political relations between countries. These benefits come at a cost, most obviously from reductions in applicable withholding tax rates and limitations on tax jurisdiction. Moreover, it is increasingly recognized that tax treaties are also a frequently used instrument for aggressive tax planning by companies and individuals.
The cost of these treaties is particularly important for recipients of capital in developing countries, which also often have limited resources to negotiate and administer tax treaties effectively. We tried to develop an approach that could help inform tax treaty policy makers in better balancing revenue and investment climate considerations. More specifically, we attempted to get a sense of both the direct and indirect costs of tax treaty policy in Ukraine by combining relevant macro and micro data. Our study attempts to estimate the tax sensitivity of aggregate income flows as well as the sensitivity of reported profitability at the firm level to changes in the relevant treaty networks.
Our results point towards a few potentially important observations. First, revenue costs linked to treaty restrictions on taxing rights are particularly relevant with respect to flows into a small number of major investment hubs – in particular, Cyprus, Netherlands, Switzerland and Luxembourg. The more attractive a treaty is in restricting taxing rights, the more revenue flows into these hubs. As World Bank Governance Global Practice Director Jim Brumby and IMF Deputy Director of Fiscal Affairs Michael Keen have noted, “tax treaties are like a bathtub; a single leaky one is a drain on a country’s revenues.”
High elasticities of income flows, however, also suggest that changes at the individual treaty partner level would not necessarily result in more revenue collection. If withholding rates are increased for a particular treaty with a partner country, income flows are likely redirected to take advantage of another attractive treaty. A comprehensive treaty policy and negotiation strategy is thus needed to address concerns on treaty related revenue losses.
Moreover, reported earnings of multinational enterprise affiliates in Ukraine suggest that the ownership structure explains reported profitability. In other words, firms linked to affiliates in certain countries with attractive treaties in place tend to be less profitable. Tax administrators looking at international accounting tactics, such as transfer pricing, should thus pay more attention to multinational corporations linked to headquarters in those countries.
Our findings point towards the importance of developing a comprehensive treaty policy approach to guide negotiations and/or revisions of tax treaty. Consequently, analysis of benefits needs to be undertaken for the treaty network as a whole. Here, we need to keep in mind that where investment activity is merely redirected via another country to take advantage of treaty benefits – but would have taken place irrespective of the treaty – revenue losses are not offset by any potential gains.