Featured Articles

Destination-Based Taxation in the House Republican Blueprint

Posted on February 17, 2017 by Wei Cui


Wei CuiWei Cui is an associate professor at the Peter A. Allard School of Law at the University of British Columbia, Vancouver. He is grateful for comments on an earlier draft by Mindy Herzfeld, Ed Kleinbard, and Peter Merrill, but all errors remain his own.

In this article, Cui examines the destination-based cash flow tax in the House Republicans' blueprint for tax reform. He addresses the likely impact of the proposed tax and some of the theoretical controversies that it has generated.

Copyright 2016 Wei Cui.
All rights reserved.

* * * * *

A month ago, the House Republican Task Force on Tax Reform released its blueprint for tax reform,1 at the center of which is a destination-based cash flow tax (DCFT) to replace the current federal income tax on corporations. To scholars of international taxation, this is a fascinating development, because the DCFT on business entities has been advocated in the last decade by some of the most sophisticated economists in the field.2 It was outlined as a tax reform option in the President's Advisory Panel on Federal Tax Reform in 20053 and has attracted the attention of policy analysts in the United Kingdom, Canada, and elsewhere.4 Yet the House GOP blueprint represents the first time that the tax has been promoted by political leaders. Initial commentators have stressed the capacity of the tax (if adopted in the United States) to reduce U.S. companies' incentives for international tax planning and profit shifting, and to allow the United States to "leapfrog to the front of the pack" in tax competitiveness.5

In this article I discuss several issues that are crucial for understanding the DCFT.6 To me it is significant not only that an intellectually stimulating idea has been given political reality, but also that even just in blueprint form, the U.S. tax reform proposal offers several implementation details that the scholars advocating the DCFT had not previously sketched out. For example, the blueprint suggests that business losses under the DCFT be carried forward with an interest factor, instead of generating a refundable tax credit.7 It also states that the DCFT would apply to noncorporate entities such as sole proprietorships and partnerships.8 Last but not least, the DCFT is presented in the blueprint as part of a package that contains a policy framework for taxing individuals. Not only is examining the DCFT in light of these implementation details of the blueprint necessary for assessing the likely impact of the proposed business tax reform, it may also help clarify some of the theoretical controversies that the DCFT has generated.

As many commentators have stressed,9 understanding the border adjustments required by the DCFT is key to evaluating its viability as well as its attractions relative to other reform proposals, such as introducing a federal value added tax (VAT) in the U.S. I start off by examining this aspect of the DCFT, presenting the problems others have identified in a different form. I argue that we should not see the "perennial question" concerning the DCFT as being about whether it violates WTO agreements. Instead, the question should be whether we truly understand the DCFT's potential impact on trade, aside from WTO legal concerns. I identify a couple of ways in which objections that can be raised against the DCFT have arguably not been adequately answered. I then examine how the loss carryforward aspect of the blueprint interacts with the DCFT's border adjustments and how things get even more complex when noncorporate entities are also taxed on a destination basis. The implementation issues for the DCFT I am highlighting would not arise if the United States were to adopt a VAT instead. Therefore, I conclude by comparing the DCFT with the VAT on the issue of progressivity and considering how other countries would respond to a U.S. DCFT.

Does the Tax Create Trade Distortions?

Much of the U.S. policy discussion regarding the DFCT's border adjustments gives the impression that a potentially very good U.S. reform is being hampered by an arbitrary legal rule imposed by the WTO: Border adjustments are permitted for "indirect" taxes like the VAT but prohibited for "direct" taxes. The rationale for the WTO rule is rarely explained. Because few believe that a distinction with any substance can be maintained between direct and indirect taxes, it seems completely fortuitous for any tax reform proposal to qualify as an indirect tax. Under that view, we are confronted with another instance in which lawyers erect senseless barriers to sensible policy -- or, worse, another instance in which the United States is bound by an arbitrary international rule imposed by foreigners.

But perhaps that is not the most helpful way of presenting the DCFT's WTO-compatibility problem.10 Consider a simple example illustrating the border adjustment aspects of the DCFT.11 Suppose that Y Corp., incorporated and operated in the United States, produces widgets and exports them to Canada. In the terminology of both the VAT and the DCFT, the United States is the country of origin and Canada is the country of destination. Suppose that each unit of the widget sells for $100 (before taking into account the goods and sales tax that Canada imposes on imports12). Assume that this price reflects a cost of $50 of wages that Y pays to its employees and $20 of a necessary intermediate service, which Y acquires from another U.S. domestic producer, X Corp. Moreover, suppose that X incurs a labor cost of $10 in producing the intermediate service sold to Y and no other cost. With these simple assumptions, X has a cash flow profit of $10, and Y has a cash flow profit of $30. In the example above, rows 1-3 display these stipulated facts. Row 4 illustrates the tax bases in the United States under an origin-based cash flow tax for X, Y, and the two in the aggregate.13Row 5 illustrates the respective tax bases under a destination-based VAT, and row 6 illustrates the tax bases under a DCFT.

Under an origin-based cash flow tax, X and Y's aggregate tax base in the United States is $40, the sum of their profits. Under the VAT, because the production of X and Y ultimately ends in exported goods, their aggregate tax base in the United States is $0. This result is achieved by Y not including the exported sales in its tax base but still deducting the cost of input purchases (but not labor cost).14 Note that even though Y gets to deduct the $20 cost of input purchase, the $20 has already been included in the tax base of X, Y's supplier. The aggregate tax base of X and Y in the United States is thus zero, not negative. That means there is no subsidy provided to the export: The price of export to Canada is simply free of any U.S. tax.

Finally, under the DCFT's border-adjustment mechanisms, Y would exclude its $100 export sale to Canada from its U.S. tax base, while at the same time deducting $70 of (labor plus other input) costs. It thus has a negative tax base of -$70. The aggregate tax base of X and Y in the United States is also negative, -$60, which corresponds to X and Y's total wage payments. If the DCFT rate is 20 percent, the U.S. government will effectively provide a $12 tax benefit ($60 x 20 percent) for the export.


              Example: Tax Base Under Three Alternative Taxes


                                           Domestic                  Aggregate
                                           Supplier X   Exporter Y   of X and Y

 1  Sales                                     20           100          120

 2  Cost of input purchase                     0            20           20

 3  Labor cost                                10            50           60

 4  U.S. tax base under an origin-based
    cash flow tax                             10            30           40

 5  U.S. tax base under VAT                   20           -20            0

 6  U.S. tax base under the DCFT              10           -70          -60

 7  Canadian tax base under the GST/VAT
    or a DCFT                                  0           100          100

To complete the picture, row 7 shows that in the country of destination, Canada, the import from Y falls within Canada's GST tax base15 (and the base of a Canadian DCFT if Canada were to adopt one as well). When Canada and the United States are considered together, X and Y's production gives rise to a tax base of $100 under the VAT ($0 in the United States and $100 in Canada) but a tax base of $40 under the DCFT (-$60 in the United States and $100 in Canada). The difference reflects the narrower tax base of the cash flow tax, which, unlike the VAT, allows the deduction of labor costs.16

The Example suggests that the option that generates the most U.S. tax for the export to Canada is the origin-based cash flow tax (which, when businesses' capital outlays and financial transactions are put aside, operates just like the current corporate income tax). The export would be subject to less -- more precisely, zero -- tax under a VAT if the United States were to adopt one. And if the United States were to adopt the DCFT, the export would actually receive a subsidy, if the negative tax base gives rise to a tax benefit.17 The subsidy would seem to raise an immediate concern for U.S. trading partners. Under the DCFT, not only would Y recover all previous tax borne by its nonlabor inputs, thus ensuring that the material cost is $20 and no more, it should also get a grant for its labor cost. If the U.S. DCFT rate is 20 percent, Y would get a $10 grant from the U.S. government per unit produced, implying that Y's net production cost is only $60. The magnitude of the grant would be even larger if the proportion of labor cost in the exported widget increases. Intuitively, that seems unfair to U.S. trade partners as well as distortionary (welfare-impairing).18 A U.S. DCFT can also have the effect of import tariffs, with the magnitude of the tariff increasing as the proportion of labor cost in the imported goods increases.19

With that in mind, the WTO's prohibition on border adjustments for direct taxes may make more sense than the U.S. policy discussion has acknowledged. The Example identifies real differences among the VAT and the origin-based and destination-based cash flow taxes. Labels about direct and indirect taxes are irrelevant.

DCFT advocates, however, have generally dismissed that line of reasoning. The first explanation they tend to offer is that the only effect of any import tariff or export subsidy (that is, not just implicit tariffs and subsidies such as one finds in the DCFT) is that it changes real exchange rates, if one assumes balanced trade in the long run.20 That is, any export subsidy or import tariff would be neutralized in the long term by exchange rate adjustments.21 That is an important idea to learn from economists, but it is not clear that it offers a full answer to potential concerns about the DCFT's trade effects. All examples that seek to show that tariffs (or export subsidies) can be neutralized through exchange rate adjustments assume that the tariffs and subsidies are imposed or granted at a uniform ad valorem rate. That assumption ensures that the relative prices of a country's exports (or imports) are not affected by border adjustments -- which therefore can be countered by a single exchange rate adjustment. But what if the border adjustments are not made at a uniform rate? As the Example shows, the amount of subsidy Y receives for its widget export depends on the proportion of labor costs in the product. Different producers thus could receive different amounts of subsidies for the same product, and different sectors would be subsidized to different extents, depending on their labor usage. At best, one could see an exchange rate adjustment countering the average subsidy a country offers to its exported products (or the average tariff imposed on imports). Even after the adjustment, however, some U.S. products and industries will enjoy residual, above-average subsidies, and others will gain an advantage over imports that face above-average tariffs. U.S. trade partners may be concerned about those products and industries, even though other U.S. products and industries (facing below-average subsidies and competing with imports subject to below-average tariffs) would be penalized by the exchange rate adjustment.

Nonuniform tax rates matter a lot when assessing the efficiency of border adjustments. In academic studies of the VAT, for example, many economists have examined whether destination- and origin-based VATs are equivalent, an idea that involves similar intuitions.22They have concluded that the most fundamental reason why the equivalence fails to hold is that real-world VATs do not tax all commodities at the same rate (because of the prevalence of VAT exemptions). Consequently, the choice between destination- and origin-based VATs must rest on a range of pragmatic (and uncertain) policy judgments, with the destination-based VAT being preferred in practice.23 Given that the acknowledgment of nonuniform tax rates has so importantly shaped policy analysis in the VAT area, it seems insufficient for DCFT proponents to reply to questions about the proposed tax's trade effects simply by pointing to a textbook theorem with demonstrably invalid assumptions.

DCFT proponents have occasionally suggested a different mechanism by which the effect of export subsidies may be neutralized in the long term. Perhaps the subsidy that the DCFT offers to labor employed in producing exported goods and services may cause the wage in the export sector (and those sectors that supply to the export sector) to rise.24 That is, in the example given above, the wage subsidy would cause X's cost of labor to rise from $50 to $60, and X thus would ultimately suffer losses and exit from the market notwithstanding the export subsidy. In effect, the suggestion seems to be that labor claims the entire benefit of the subsidy or grant for wage payments -- that is, it bears the full incidence. However, the theoretical and empirical validity of that assumption about the incidence of a tax benefit on labor is again likely to be controversial. Without justifying those assumptions, it is not clear how DCFT proponents can dispel concerns about distortionary trade impacts.

In summary, even in the absence of WTO legal prohibitions, distortionary trade subsidies presumably should be viewed as undesirable, just like distortions of corporate decisions on locations of production and the intensity of capital investments. Therefore, the WTO law should be seen not as a foreign obstacle for the DCFT but as a relevant critique in economists' own terms. Many economists and policy analysts have objected to the DCFT on the grounds of its WTO incompatibility.25 If the WTO rule was purely arbitrary, perhaps this controversy would not have remained so "perennial."

Loss Carryforward Under the Blueprint's DCFT

In the House GOP blueprint, the net operating losses of a business are carried forward, with an interest factor to compensate for inflation and to ensure that the normal rate of investment return is not subject to tax.26 Thus, generally, net losses in the current period can offset future earnings but do not lead to any cash transfers from the U.S. Treasury. Yet for an exporting business, one presumably has to allow current deductions of costs to offset other sources of the business's taxable profit because the business's revenue from exports will always be excluded from the tax base. Therefore the export subsidy (shown in the Example) comes in the form of tax benefits reducing the tax liability on the business's other taxable profit. By contrast, the issue does not arise under the destination-based VAT (row 5 of the Example) because any rebate an exporter (such as Y) gets under the VAT corresponds only to the VAT that it has been charged on its input purchase (and which has been paid to the vendor, such as X, who in turn remits it to the government).

The blueprint's loss carryforward approach is an area in which the legislative proposal has clarified the past, purely academic discussions of destination-based taxation. Economists who advocate the cash flow tax (or other similar taxes on corporate rent), whether of the destination- or origin-based variety, have traditionally insisted on a full cash subsidy for corporate losses. They have done so for two reasons.27 First, the asymmetrical treatment of profit and loss resulting from risk taking (that is, profits from lucky outcomes are taxed but losses from unlucky outcomes are disregarded) discourages risk taking. Second, it is difficult to distinguish economic rent from returns to risk taking for particular businesses and investments. When a business realizes an outsized return, it is generally hard to say how much was the result of the company seizing on a unique opportunity and how much was just good luck (it is usually both). Only by taking full account of losses in the tax system -- by allowing full offset of losses against profit and the refund of negative tax liabilities of individual businesses -- can one address those two problems.

However, almost no real-world tax systems offer refundable tax credits or cash subsidies for losses. The reason is not mysterious: It is much easier to lose money than to make money, and any government should be loath to partake in all the loss opportunities out there. This approach is true not only of the income tax but also of the VAT. In this regard, it is easy to be misled by the fact that for a particular firm, it is possible for VAT input tax credits to exceed VAT payable on sales, with the result that the company gets a VAT refund depending on the excess of its cost of input purchases over its sales. One should not forget that this refund is for the VAT that has previously been charged to the company on its input purchases. A VAT refund simply ensures that no tax is collected in excess of the value of a company's taxable sales. It does not require the government to offer a subsidy to any business when there is a loss. (Any tax revenue refunded had literally come out of the taxpayer's own pocket.) To put it differently, the VAT taxes consumption even if the consumption is produced through processes generating net losses.28

In other words, there was a large gap between the cash flow tax economists wrote about (in both theoretical and even policy papers) and real-world taxes. If the GOP blueprint's brief statement about loss carryforwards is read to apply to exporting businesses, it can be seen as beginning to close that gap by eschewing cash grants for losses.29 Nonetheless, there still appear to be some intolerable gaping holes. It seems to follow from the design of the DCFT that unless "losses" recorded by exporters can be freely traded for cash with other businesses,30 no corporation would want to be engaged purely in exports because its negative tax base would not result in any tax benefit. All exporters would want to acquire businesses generating domestic sales or enter into group consolidation with corporations that have those sales. That type of behavior would often be distortionary -- they would not occur in the absence of the potential tax benefits enjoyed by exporters. Moreover, if different businesses use their losses from exports to different extents, depending on how much taxable profit their domestic sales generate, that could introduce another dimension of heterogeneity -- as well as fluctuation -- to the amounts of subsidies that different exporters receive.31

The problem of loss utilization would not be nearly as dramatic under an origin-based cash flow tax that also adopts the loss carryforward with interest. A business's sales would always be included in that tax's base. An interest factor attached to losses carried forward would often be adequate to preserve the value of losses realized, except when a company liquidates without earning a profit. The nonrefundability of companies' terminal negative tax liability still deters risk taking to an extent, but that is a matter of degree and is not different in kind from the treatment of losses under the current income tax (which theorists view as not ideal). By contrast, there has to be some way to give effect to the negative tax liability associated with all exporters under a DCFT; otherwise it would not be a destination-based tax. But the choices among direct cash grants for current excess deductions, costless (safe harbor) loss trafficking, and more limited loss utilization through tax planning all seem to be unpalatable, as well as unprecedented.

Taxing Noncorporate Entities by Destination

"The cash flow based approach that will replace our current income-based approach for taxing both corporate and noncorporate businesses will be applied on a destination basis."32 That statement in the GOP blueprint confirms what seems to be an inevitable implication of DCFT proposals, namely that it must be applied to corporations and noncorporate business entities alike. As a practical matter, it is hard to imagine corporations being taxed on a destination basis but noncorporate entities taxed on an origin basis. If corporations are not allowed to deduct the cost of imports but partnerships are, imports would all be done through partnerships. Partnerships would then on-sell the imported goods to corporations, defeating the tax on imports. Conversely, all noncorporate entities making export sales -- think of law, accounting, consulting, and financial management firms providing services to foreign clients -- would all try to make those sales through corporations, if corporations can exclude export sales from their tax base but noncorporate entities cannot. Moreover, corporations entering into joint ventures that involve selling goods and services across borders would have to rethink whether they would want to do so using noncorporate vehicles.

Although applying the DCFT to corporations makes it practically necessary to apply the same treatment to noncorporate entities, there does not otherwise seem to be a motive for that fundamental change to the taxation of noncorporate entities. Because individual owners of sole proprietorships, partnerships, etc., are taxed on the entities' income, American taxpayers cannot use them to engage in deferral planning. Also, because of the United States' foreign tax credit regime and because individuals cannot manipulate their tax residence as easily as corporations can, distortions of residence- and source-based taxation either do not arise or are minimal. Therefore, international tax planning and profit shifting by noncorporate entities taxed on a passthrough basis are not issues of serious concern. Consequently, although cash flow taxation makes a difference to all businesses because of the benefit of immediate expensing (and the detriment of interest nondeductibility), the destination aspect of the DCFT introduces no efficiency gain in the treatment of passthrough entities. Indeed, according to other tax reform proposals advanced in the United States, the corporate tax system should be reformed to align more with the regime for passthrough taxation, not the other way around.33

One may even question whether the destination-based tax treatment of noncorporate entities is positively base eroding (as well as regressive). Under the DCFT, U.S. lawyers, business consultants, investment managers, and others providing professional services to foreign clients and offshore companies would no longer be taxed on the profits earned from those services -- not even under the reduced 25 percent rate that the GOP blueprint proposes for passthrough entities' active business income. Instead, they would be taxed on those profits only when, and to the extent, they finance U.S.-based consumption from those profits.34 Moreover, those companies could have negative tax liabilities after the deduction of wage costs. The consequence may be that instead of looking for loss-generating tax shelters as they did in the past, U.S. high-income-earning service professionals would look to acquire interests in businesses generating active income in order to make use of their export-related deductions. That adds another layer to the loss trafficking problems generated by corporations discussed in the previous section. It may also exacerbate preexisting incentives to convert wages taxed at a higher rate (for example, the 33 percent maximum under the blueprint) to active business income (which can be subject to zero tax if offset by loss carryforwards).

The preceding discussion assumes that under the DCFT, noncorporate entities would compute profits or losses in the same way as corporations but would nonetheless not be subject to the DCFT at the entity level. Instead, the owners of noncorporate entities would be subject to tax (at a maximum 25 percent rate, according to the GOP blueprint).35 But that is arguably an incorrect way of thinking about things. The defining feature of passthrough entities is that their owners are taxed on the entity's income. Corporations, by contrast, are subject to an entity-level tax, separate from shareholder-level taxation of dividends and capital gains. However, this distinction makes sense only in the income tax context: Because corporate shareholders enjoy the benefit of deferral (that is, no current inclusion), an entity-level tax is needed to undo that benefit. However, under any type of cash flow tax applied to corporations (including the DCFT), the entity-level tax would no longer have the effect of undoing deferral because cash flow taxes allow the immediate expensing of all of a company's capital investments. Companies would thus always earn a normal investment return without being subject to tax and would be indifferent between paying tax earlier or later. Individual shareholders would enjoy the benefit of perfect deferral on investments in a company subject to the cash flow tax.

That implies that if the DCFT were adopted, it would be artificial to maintain the distinction between corporate and passthrough entities -- taxing the former both at the entity and owner levels and taxing the latter only at the owner levels. Indeed, none of the past cash flow tax (or equivalent) proposals advanced in the United States took that approach. David Bradford's "X tax," for example, is simply a consumption tax that tries not to tax shareholders on corporate income.36 More recently, Edward D. Kleinbard proposed the business enterprise income tax (BEIT), which explicitly taxes only economic rent at the company level and only the normal return to capital at the investor level.37 The BEIT serves as a device to measure returns to capital from risk taking and economic rent, instead of the traditional role of preventing the deferral of income by shareholders. Under both Bradford's and Kleinbard's proposals, owners of corporate and noncorporate entities are taxed alike, as are the two types of entities themselves.38

By contrast, under the House GOP blueprint, owners of corporate and noncorporate entities continue to be subject to different treatments (that is, two levels versus one level of taxation) even though that is no longer justifiable in terms of the entity-level tax that corporations are subject to. An alternative way of implementing the DCFT should therefore be to tax owners of noncorporate and corporate entities alike: Partners, for example, should be taxed on both distributions and sales of partnership interests, while the 20 percent DCFT would be imposed at the partnership level. That would at least improve one aspect of the U.S. tax system, by eliminating the complexity of passthrough taxation.

Progressivity and International Response

The House GOP blueprint asserts that the DCFT would "allow the United States to adopt, for the first time in history, the same destination-based approach to taxation that has long been used by our trading partners. This will end the self-imposed unilateral penalty for exports and subsidy for imports that are fundamental flaws in the current U.S. tax system."39 The comparison made here is puzzling. Most other OECD countries impose both an origin-based corporate income tax and a destination-based VAT. Although their corporate income tax rates are generally lower than those of the United States (and may be more "competitive" in other ways), it is their VATs, not their corporate income taxes, that allow for border adjustments. To say that the DCFT "counters" U.S. trade partners' VATs implies that VATs involve export subsidies and ignores that other countries maintain origin-based corporate income taxes. Neither is correct (as can be seen from the Example).40Whatever self-imposed penalty that is implicit in the origin-based corporate income tax is shared by the United States and its trade partners and is thus by no means unilateral. What distinguishes the United States is its resistance to adopting the VAT.

To put it differently, if the United States simply were to replace its corporate income tax with a destination-based VAT, the change that would matter to the rest of the world would not be the United States eliminating a self-imposed penalty (previously failing to border adjust under an income tax). Instead, it would be that the United States would have eliminated its corporate income tax, while others still maintain those taxes in addition to VATs. As far as the taxation of multinationals is concerned, it is hard for a country to be more competitive than that.41 By the same token, that would certainly end the phenomenon of U.S.-tax-driven multinational enterprise tax planning. The business competitiveness goals of the GOP blueprint would all be thoroughly achieved -- and, as it happens, the VAT would also be WTO-compatible and free of the uncertainties regarding implementation discussed above.

The new question here is not why U.S. politicians would not advocate adopting the VAT (even though the House GOP now seems willing to consider the DCFT). It is instead why the DCFT's intellectual proponents would favor it over the VAT. One suggestion is that the DCFT may be more progressive than the VAT.42 The reasoning behind that suggestion is probably the following: The VAT taxes all consumption, whether the consumption is financed by wage earnings or supranormal returns to capital. The DCFT, by contrast, allows wage payments to be deducted from the tax base, and economists have concluded that this means the DCFT would tax only consumption that is financed out of supranormal returns to capital investment.43 If higher-income individuals finance a greater proportion of their consumption from returns to investment rather than from wage earnings, perhaps a tax on investment-financed consumption has a built-in progressivity. (Because money is fungible, presumably a higher proportion of investment-financed consumption simply means a higher proportion of current disposable income that comes from wealth instead of labor.)

That type of consideration could seem relevant when one considers only business tax reform -- which is what the DCFT's intellectual proponents have focused on.44 However, its relevance is greatly diminished when included in a reform package that also addresses individual taxation. That is precisely what the blueprint has done. According to the Republican proposal, investment income (dividends, capital gain, and interest) would all be taxed at progressive rates, although effectively at half the rates of other types of income in the same brackets (through a 50 percent deduction). That seems to be a far more direct and effective way of introducing progressivity than modifying the VAT. Whatever progressivity there is in the DCFT relative to the VAT seems to lose policy significance.

Recognizing that the United States is unique among OECD countries in not having a VAT is also relevant to the question of how other countries might respond to a U.S. DCFT. A basic reason why that question is unanswerable is of course that the DCFT may be WTO-incompatible, but even putting the WTO aside it is unclear whether countries could come to terms with the trade effects of the DCFT. Moreover, as was suggested above, the export subsidies required by the DCFT imply mechanisms for governments to transmit cash to their exporters in ways that are unprecedented in real-world tax systems. But suppose that those problems are somehow overcome. Suppose, for example, that the United States simply allowed unlimited trafficking of losses arising from labor cost deductions from exported sales. Would other countries follow suit and adopt the DCFT as well?

My guess is that the answer is no. The reason is that the DCFT is too much like the destination-based VAT. The only differences are the deduction for labor costs and the fact that the DCFT operates on a subtraction as opposed to a credit-invoice method. Other countries already have the VAT, and there is little justification to impose a separate subtraction-type tax that is also destination-based. To achieve the effect of the DCFT, they would only have to increase their VAT rates (by the rate of the desired DCFT), while at the same time adopting some mechanism to give effect to labor cost deductions. For example, they could offer a deemed input VAT credit (at the rate of the desired DCFT) for all labor costs incurred by VAT taxpayers. That could undermine the integrity of their VAT systems in the way loss trafficking would undermine the integrity of income tax systems. But at least they would not have to worry about the possibility that their (personal) income tax systems would be undermined by a substantial new source of business losses.


1 Tax Reform Task Force, "A Better Way: Our Vision for a Confident America" (June 24, 2016) . For initial reports and commentaries, see, e.g., Martin A. Sullivan, "Border Adjustments Key to GOP Blueprint's Cash Flow Tax," Tax Notes, July 18, 2016, p. 303; Ryan Finley, "Replace Corporate Tax With a Cash Flow Tax, Economists Say," Tax Notes, July 18, 2016, p. 330; Kyle Pomerleau and Stephen J. Entin, "The House GOP's Destination-Based Cash Flow Tax, Explained" (June 30, 2016), available at http://taxfoundation.org/blog/house-gop-s-destination-based-cash-flow-tax-explained; Pomerleau, "Details and Analysis of the 2016 House Republican Tax Reform Plan" (July 5, 2016), available at http://taxfoundation.org/article/details-and-analysis-2016-house-republican-tax-reform-plan; and Dylan F. Moroses, "Tax Foundation's Score of Tax Reform Blueprint Questioned," Tax Notes, July 11, 2016, p. 177. This article was prepared in July 2016, before subsequent commentaries published in Tax Notes. But nothing has changed the analysis that follows.

2See Alan Auerbach and Michael Devereux, "Consumption and Cash-Flow Taxes in an International Setting," National Bureau of Economic Research working paper No. 19579 (Oct. 2013). There is also an important body of U.S. literature that discusses implementing consumption taxation through a cash flow tax on businesses, either on a destination or an origin basis. See, e.g., David Bradford, "The X Tax in the World Economy," Griswold Centre for Economic Policy Studies working paper No. 93 (Aug. 2003), available athttps://www.princeton.edu/ceps/workingpapers/93bradford.pdf; and Harry Grubert and T. Scott Newlon, "The International Implications of Consumption Tax Proposals," 48 Nat'l Tax J. 619 (1995).

3See President's Advisory Panel on Federal Tax Reform, "Simple, Fair, and Pro-Growth: Proposals to Fix America's Tax System," at ch. 7 (Nov. 2005).

4See Alan Auerbach et al., "Taxing Corporate Income," in Dimensions of Tax Design: The Mirrlees Review 837-893 (2010); Robin Boadway and Jean-François Tremblay, "Corporate Tax Reform: Issues and Prospects for Canada," Mowat Centre research paper No. 88 (May 7, 2014) , available at http://mowatcentre.ca/corporate-tax-reform; and IMF, Spillovers in International Corporate Taxation 42 (2014).

5 Sullivan, supra note 1, at 304-305.

6 For a more extended discussion of different varieties of DCFT proposals and their relationship to the OECD's base erosion and profit-shifting project, readers are referred to my working paper, Wei Cui, "Destination-Based Cash Flow Taxation: A Critical Appraisal" (Sept. 30, 2015), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2614780.

7 Tax Reform Task Force, supra note 1, at 26.

8Id. at 28.

9See Sullivan, supra note 1, at 304-305; Wolfgang Schön, "Destination-Based Income Taxation and WTO Law," Max Planck Institute for Tax Law and Public Finance Working Paper 2016-03.

10 The following discussion is agnostic about the intent, logic, or structure of that aspect of WTO rules, of which I am not an expert.

11 The Example replicates the cost structure of the chain of production described in the example in Sullivan, supra note 1, which he uses to illustrate the import tariff effect of the DCFT.

12 The GST is the Canadian version of the VAT.

13 This Example does not involve capital outlays or borrowings of either company, and therefore the cash flow profits of both are as they would be under the income tax.

14 Under the invoice-credit VAT adopted by most countries, that result would be achieved by "zero-rating" exports and offering refunds of input VAT previously charged to Y.

15 Although this is usually collected from and even nominally charged to the importer in Canada, the point is that Y's export sale is taxable in Canada.

16 The origin-based cash flow tax (shown in row 4 of the Example) also has this narrower tax base, but under that tax, all net profits are taxed in the United States, the country of origin.

17 See below for further discussion of the loss carryforward mechanisms under the House GOP blueprint.

18 Suppose that the world producer price of the widget that Y exports, before capital costs (i.e., the required return to Y's capital), is $60. Y's production of the good in the United States, which incurs a cost of $70, is thus unprofitable and loses $10 per unit in a competitive market. The U.S. subsidy of $10 would allow it to break even.

19 Sullivan, supra note 1, at 304. To see this through the Example, consider what would happen if Canada, the country of destination, were to impose a DCFT on the full $100 of import from Y. If an independent chain of production and sale similar to the one depicted in the Example took place purely domestically in Canada, then, like row 4 depicting the origin-based cash flow tax, Canada would have a tax base of $40. (Origin- and destination-based taxes are identical in effect for purely domestic transactions without imports and exports.) It follows that imported goods would be taxed more heavily in Canada under the DCFT and would have the effect of an import tariff.

20 For a good popular explanation of that idea, see Alan D. Viard, "Keynes at the Border?" American Enterprise Institute (Apr. 15, 2009), available athttp://www.aei.org/publication/keynes-at-the-border/. For a more detailed explanation, see Jagdish N. Bhagwati et al., Lectures on International Trade 215-219, section 12.6, "Lerner Symmetry Theorem" (1998).

21Seee.g., Bradford, supra note 2; and the transcript of the Conference on "Tax Reform in an Open Economy," held at the Brookings Institution and organized by the Urban-Brookings Tax Policy Center and International Tax Policy Forum (Dec. 2, 2005).

22Seee.g., Lockwood et al., "When Are Origin and Destination Regimes Equivalent?" 1(1) Int'l Tax & Pub. Fin. 5 (1994).

23Seee.g., Michael Keen and Walter Hellerstein, "Interjurisdictional Issues in the Design of a VAT," 63 Tax L. Rev. 359, 363-367 (2010). This may be seen as the obscure and technical intellectual justification of the WTO's permission for VAT border adjustments.

24See Kristen A. Parillo, "A Destination-Based Corporate Tax: An Alternative to BEPS?" Tax Notes Int'l, Apr. 27, 2015, p. 315, at 320 (quoting Michael Devereux as responding to the objection that the DCFT creates an export subsidy by claiming that "prices would adjust -- just like under a VAT").

25See Bradford, supra note 2, at 12-13; and Boadway and Tremblay, supra note 4, at 47.

26 Tax Reform Task Force, supra note 1, at 26.

27Seee.g., Bradford, supra note 2; and Boadway and Tremblay, supra note 4. For recent reflections on whether the traditional approach is justified, see Boadway et al., "Cash flow Business Taxation Revisited: Bankruptcy, Risk Aversion and Asymmetric Information," Oxford Centre for Business Taxation working papers 15/31 (2015).

28 Cash flow tax advocates arguably miss this point when they claim similarities -- but for the deduction for labor costs -- between real-world VATs and the tax they favor in theory. For example, Bradford suggested that under a VAT, any investment outlays are immediately deducted in the computation of VAT liability. As a result, "the general public shares in the investment and payoffs in proportion to the tax rate. In making investment decisions, the taxable firm considers its share." Bradford, "Consumption Taxes: Some Fundamental Transition Issues," in Frontiers of Tax Reform 132 (1996). That is incorrect: The government simply does not share the risk of business loss through the VAT.

29 The report of the 2005 President's Advisory Panel recommended (under the destination-based growth and investment tax) allowing cash subsidies for exporters with excess deductions, even though businesses making domestic sales would only be able to carry forward losses. President's Advisory Panel on Federal Tax Reform, supra note 3, at 171. It acknowledged that "special rules may be needed to police the allocation of expenses between domestic businesses generating losses and export businesses when both are operated within the same firm or through affiliates." I believe that understates the problem: The opportunities for tax avoidance and even fraud associated with cash subsidies to exporters are bad enough in themselves, even without considering the strong incentives for tax planning when exporters and domestic businesses are treated differently.

30 For the government, that would be just as bad as giving cash directly to the exporters. The report of the 2005 President's Advisory Panel acknowledges the risks of allowing loss trading under any cash flow tax: "Allowing tradable or refundable losses may encourage tax avoidance schemes in which the taxpayers make investments that would not have been worth undertaking in a no-tax setting. The value of tax losses created by such an investment may be a key component of its appeal. In addition, allowing loss trading could make it much more important to police so-called 'hobby losses' and losses generated by various forms of disguised consumption, rather than investment, because those losses could generate tax savings even when the person incurring them would never realize offsetting positive cash flow." President's Advisory Panel on Federal Tax Reform, supra note 3, at 167.

31 In this context, it would become even harder to see how (as DCFT proponents would want us to believe) the subsidy or tariff apparently present in the DCFT can be neutralized by exchange-rate adjustments or labor moving across firms and sectors. How flexible would the exchange rate have to be to adjust even on average to those fluctuations, and how would wages reach equilibrium?

32 Tax Reform Task Force, supra note 1, at 28.

33See Eric Toder and Viard, Major Surgery Needed: A Call for Structural Reform of the U.S. Corporate Income Tax, 1 (2014).

34 The fact that U.S. taxpayers can earn investment returns at a low tax rate and then migrate to a low-tax jurisdiction when they retire and consume is deemed to be a general problem for a destination-based cash flow tax (as compared with an origin-based income or cash flow tax) and is not specific to owners of passthrough entities. See Bradford, supra note 2, at 26-27.

35 This is of course inherently distortionary.

36 Bradford, supra note 2.

37 Kleinbard, "Reimagining Capital Income Taxation" (paper presented at the Annual Symposium of the Oxford University Centre for Business Taxation, Saïd Business School, Oxford, June 22, 2015). The BEIT implements the tax on corporate rent through a cost of capital allowance instead of immediate deductions for capital expenses, a distinction that is irrelevant for the purposes here.

38 The growth and investment tax considered by the report of the 2005 President's Advisory Panel applies to corporate and noncorporate entities in the same way, but owners of the two types of entities are still taxed differently. The growth and investment tax is thus vulnerable to the same criticism directed at the DCFT here.

39 Tax Reform Task Force, supra note 1, at 15. See also id. at 28 ("For the first time ever, the United States will be able to counter the border adjustments that our trading partners apply in their VATs.").

40 To refer back to the Example, U.S. trade partners all impose taxes corresponding to row 4 (except that most of them have origin-based income taxes and not cash flow taxes) and row 5 (the VAT), whereas the United States has only had a tax corresponding to row 4 (with the caveat that it is not a cash flow tax). Replacing the corporate income tax with the DCFT would introduce export subsidies (as in row 6 of the Example) not observed in row 5, which depicts the neutrality of the VAT.

41 As shown by the discussion above, however, the DCFT has a narrower tax base than the VAT. Replacing the corporate income tax with a DCFT (without having a VAT at the same time) would certainly be more business favorable than replacing it with a VAT.

42 Finley, supra note 1.

43 Auerbach and Devereux, supra note 2.

44 The DCFT's scholarly advocates have tended to be reticent on what type of individual-level taxation they assume and how the DCFT relates to that taxation.