In Thinking Outside the Tax Treaty, Professor Adam H. Rosenzweig tackles a longstanding problem for income taxation: namely, its constant undermining due to aggressive tax avoidance and evasion.1 Professor Rosenzweig envisions a better outlook for income taxation through increased voluntary participation in information sharing by countries that, under the current status quo, profit far more by remaining out of reach.2 He uses game theory to show how we might reformulate the rewards and punishments offered by the international tax regime in order to change incentives rather than try to force countries to buy in to this order against their own interests.3 In other words, Thinking Outside the Tax Treaty outlines a “best alternative to a negotiated agreement” (BATNA) for international tax.4
In this Reply to Professor Rosenzweig’s article, I seek to affirm that the instinct to look for an international tax BATNA is a good one because the international community appears to be mired in a
negotiated-agreement-or-nothing status quo, with no clear way to save the income tax from evisceration via aggressive tax avoidance and evasion. However, I am afraid that Professor Rosenzweig’s proposed BATNA will fall by the wayside as the United States appears intent on continuing the longstanding tradition of preferring negotiated agreements that use too many sticks and not enough carrots to try to control the international tax order to its own advantage.5 This would be unfortunate, because Professor Rosenzweig’s proposed BATNA has much more to offer than the status quo, both for the cause of order in international tax coordination as a general matter and, more specifically, for the plight of countries that are currently ostracized from this order.
I. Why Does International Tax Need a BATNA?
In order to understand why international tax needs a BATNA at all, it helps to consider what constitutes the essential problem that globalization poses for income tax policy and really any policy meant to apply to persons or resources that move easily across sovereign borders. Professor Rosenzweig succinctly encapsulates the essence of the problem as “a flaw in the basic assumptions underlying the analysis of the modern international tax regime . . . that cooperation is always in the best interest of all countries of the world.”6
The most common illustration of this false assumption emerges in the persistent information asymmetry that allows for global tax evasion.7 The oversimplified but conventionally accepted story is that governments seek to extract tax revenues from wealthy taxpayers. In response, these taxpayers seek to conceal information about their wealth from their governments. Other countries can exploit this situation to their benefit by offering safe shelter in exchange for administrative or management fees that represent a fraction of what would have been collected in tax revenues by the home government.8 The solution for the taxpayer’s “home” government is simply stated: it must find a way to extract the necessary information either from the taxpayer or someone else who also has it. But extracting information is anything but simple in a world in which states are contained within territorial limits while capital roams free.
Governments generally eschew chasing the taxpayer herself when she moves across sovereign borders, since that involves exerting extraterritorial jurisdiction. Until recently, at least, exercising extraterritorial jurisdiction was viewed as improper (since it necessarily interfered with another state’s jurisdiction), but also impracticable (since states have little means to enforce such claims short of physical intervention). Tax compliance therefore generally meant finding intermediaries that could serve as information repositories, and extracting useful information from them.
Governments have historically viewed their foreign counterparts as the most appropriate such intermediaries in cross-border cases, so extracting the necessary information became a diplomatic exercise, achieved by negotiating tax agreements among sovereigns.9 Where two countries have stood in relative parity, the negotiated agreement has been a friendly quid pro quo—a tax treaty that allocates taxing rights among the countries and provides for information sharing on a reciprocal basis.10 Where the two countries have not stood in relative parity because one stands to benefit more by helping taxpayers evade taxation than by cooperating, the answer has still been a negotiated tax agreement, specifically a “tax information exchange agreement” (TIEA). But in the case of a TIEA, the basis for the agreement is mainly coercion and threat of sanction (economic or political)—the quid pro quo consists only of promises not to impose penalties.11
Professor Rosenzweig expresses an instinct about this status quo that must surely be correct: the only possible outcome of this scenario is repeated failure to cooperate and therefore a perpetually broken international tax order. Countries that gain mutual benefit from cooperation will do so on friendly terms. Countries that do not stand to gain mutual benefits will enter into coercive relationships. Predictably, those that stand to lose under coercive agreements will have high incentives to defect in practice on what they agreed to in principle. Moreover, countries deemed insufficiently important to warrant the trouble of either a friendly or a coercive agreement will simply be isolated and ignored, as they generally have been for the entire history of income taxation in the United States.12The inevitable result is the information symmetry stalemate Professor Rosenzweig describes. Clearly, some alternative to negotiated agreements is both necessary and welcome to solve the coordination problem.
II. Could An Arbitration Forum Serve as a BATNA?
Once we abandon the assumption that all countries would benefit if only they could all agree to cooperate on tax matters, we can see why persistent efforts to achieve such cooperation through negotiated international agreements have been failing for so long. These efforts fail because universal agreement on tax matters would not produce any benefit for many smaller, poorer countries; indeed, universal agreement could ultimately destroy their place in the global economic order. These countries face only a series of suboptimal choices. If they do not cooperate, they face sanction. If they do cooperate, they face financial ruin. Only coercion will compel any cooperation beyond superficial promises, and it is clear that coercion must be backed by constant monitoring and credible threat of punishment for noncompliance.
One answer to this problem is to find ways to bring countries into the global order incrementally, without scaring them off with the kind of immediate full-scale, against-interest participation that negotiated agreements tend to impose. This approach is Professor Rosenzweig’s BATNA: he suggests that rather than continuously trying to bind countries to agreements that do not benefit the parties equally, it is time to consider an international forum where nations can submit disputes on a case-by-case basis about how two nations will share the tax revenues from a given income stream.13 In such a forum, a country that was not willing to be cooperative all the time (or not big enough to attract a negotiated agreement even if it were willing to cooperate) could be cooperative some of the time. This, Professor Rosenzweig suggests, would ultimately benefit the international tax system as a whole.
Professor Rosenzweig’s system proposes one monetary and one diplomatic reward for states that are currently uncooperative in the context of the information asymmetry problem.14 The monetary reward would be a dispute resolution mechanism weighted in their favor: such countries would “win” the tax revenues for most, or perhaps all of the cases brought to the tribunal, so long as they agreed to a full information exchange with respect to that dispute.15 The diplomatic reward would be a signaling effect: namely, that by complying with the forum, states would be viewed as cooperative nations, even if they did not have functioning negotiated agreements to demonstrate their bona fide intentions in the international community of states.16
I am not necessarily convinced that the proposed rewards will bring states to the table,17 For example, I am not sure that winning one disputed transfer pricing case would be seen as worth sacrificing the information that might lead the United States to pursue more of the same kind of strategies, if the country in question benefits more from providing flexible transfer pricing rules than from collecting any tax revenues on income allocated thereunder.18
Once we understand that cooperation is not going to emerge voluntarily under the status quo because it is a bad deal for too many countries, Professor Rosenzweig’s search for a BATNA makes sense. He starts from the proposition that coercion into negotiated agreements is not a viable long-term strategy once we realize that traditional tax agreements do not and cannot offer enough benefits to enough countries. Therefore the best that the international tax system can offer as an alternative is a means of trading for cooperation on a case-by-case basis, rather than continuing in a futile effort to extract it on a country by country basis.19
Professor Rosenzweig’s argument is thus for incremental cooperation. He argues that the best path to full cooperation is partial cooperation, where the alternative is not only failure to cooperate, but also active noncooperation in the form of undermining behavior (such as harmful tax competition). It follows that rich countries like the United States will benefit more in the long run by yielding and parrying under a BATNA scenario than by striking continuously with the same age-old negotiated agreement stick. In other words, giving up revenues in specific cases now in order to draw more governments into an arena where cooperation takes place is better than imposing sanctions against governments that will not cooperate with the United States’ preferred rule enforcement via negotiated tax agreements.
Unfortunately, there are reasons to be pessimistic that the proposed BATNA will be welcomed by the United States, mostly notably because of the recent enactment of the Foreign Account Tax Compliance Act (FATCA), which seeks to entrench the negotiated agreement status quo with yet more sticks but no new carrots.20 In brief, FATCA’s approach is to impose penalties on financial intermediaries unless they agree to provide information on U.S. taxpayers under a TIEA-style coercive model of no reward for compliance other than the absence of penalty.21 As Professor Rosenzweig notes, this is not a viable long-term strategy for a coherent international tax order:
When a country punishes one of its own taxpayers for investing in a tax haven, it reduces the payoffs to that taxpayer but does not punish taxpayers from other countries.
What results, therefore, is not a move to the Pareto-superior state, but rather a move to the worst case for the country’s own taxpayer and a shift of benefits to taxpayers from other jurisdictions.22
Professor Rosenzweig’s observations suggest that the continued focus on coercive strategies to extract information is potentially damaging to the global income tax order in the long run. The strategy also represents an opportunity cost, since it likely precludes any possibility for a Rosenzweig-style BATNA regime to emerge as a test case. That is a real loss because the BATNA Professor Rosenzweig envisions is one that moves toward bringing more countries into a coherent order—especially those that are smaller, poorer, and economically dependent on attracting foreign capital. Many of these are the very countries that stand to lose the most from the perpetuation of the status quo defective international income tax order.23
Finding a BATNA for international tax is a novel idea for observers of global tax policy, for whom negotiated agreements have long been cast as the only possible means of achieving coordinated action in income taxation. Tax policy makers and scholars have viewed negotiated tax agreements in the aggregate as the main, if not sole, avenue to anything like an international tax system. This has been true ever since the first group of experts got together to talk about the problem of overlapping income tax regimes and came up with bilateral agreements as the second best alternative to the optimal result: namely, a multilateral treaty that would carve up the world’s resources and its peoples and assign them to a single taxing authority. Continuing in that tradition, even when the primary aim is to twist the arm of an unwilling counter party, the mechanism of choice has been negotiated agreements.24 In Thinking Outside the Tax Treaty, Professor Rosenzweig provides a welcome suggestion for a third way between coerced cooperation and noncooperation. His proposed BATNA might allow governments the policy space to reflect upon and reconfigure their goals for international income taxation.
* H. Heward Stikeman Chair in Tax Law, McGill University Faculty of Law.
WLR Online: Volume 2013, No. 2