Will House Republicans’ “Border Adjustable” Tax Plan Cause a Trade War? (Spoiler: Maybe Not!)

Perhaps the highest legislative priority for House Republicans in the 115th Session of Congress is an overhaul of the United States’ antiquated and onerous corporate tax code. The details of the GOP plan aren’t out yet—it’s only summarized in a House Republican “blueprint” and the legislative text hasn’t been finalized—but the general idea is to replace the current 35% “worldwide” corporate income tax with a 20-25% “destination-based” tax on corporations’ US sales (i.e., including domestic sales of imports in the tax base, but excluding export sales—so called “border tax adjustments”).

This “destination-based cash-flow tax” (or “#DBCFT” as the tax nerds are now calling it on the interwebs) would be a fundamental shift in how (and on what) American corporations now pay tax, and it raises complex economic issues—including how trade would be affected—that I wouldn’t dare try to navigate here (but these analyses are a good start). 

On the other hand, the DBCFT debate also has addressed whether the tax plan would be consistent with World Trade Organization (WTO) rules on “border adjustable” taxes, with some folks already going so far as to claim that any version of the DBCFT would violate the United States’ international obligations and thus expose US exports to billions of dollars-worth of WTO-sanctioned retaliation. As I discuss below, however, that assumption’s not really correct; in fact, there is a very good argument that the DBCFT, if properly constructed, would pass muster at the WTO and thereby avoid potential retaliatory tariffs on US exports.

Before we get to that analysis, however, three very important notes:

  • First, the Republican plan is not a tariff or other border measure that applies only to imports; it is an “internal tax” that applies to both imports and domestically-produced goods and services. (Think of it like a sales tax, which is paid on a good or service consumed in the United States, regardless of its origin.) This has two important implications: 1) the tax is not the same as, for example, a steep “border tax” on imports into the United States from US companies that engage in “outsourcing” (though many have speculated that the Republican tax proposal could serve as a final compromise between Congress and President Trump on the issue of “outsourcing” and border taxes); and 2) the tax plan is therefore not inherently “protectionist” (though it may have important distributional effects, depending on things like currency movements—see economic links above).
  • Second, and relatedly, WTO rules are primarily concerned with preventing discrimination in a Member’s domestic market—in favor of certain WTO Members’ goods and services (the “most favored nation” principle) or domestic industry (the “national treatment” principle or rules prohibiting certain subsidies). Thus, a WTO Member is relatively free to apply many types of measures, such as the DBCFT, that could have significant effects on trade flows or domestic investment but might not actually violate those WTO rules. And, whether other WTO Members reciprocate with similar taxes on US-origin goods and services is immaterial to any ultimate conclusions of the DBCFT’s WTO-consistency (though it may have a practical effect). These distinctions might sound wonky (and they are), but they’re important: while most of the trade-related focus on the DBCFT has been on the taxation of imports and rebate or exemption for exports or on whether other WTO Members apply similar taxes to American products, the WTO focus is, as you’ll see below, instead on two things: 1) whether the tax itself discriminates against imports in favor of identical domestic goods in the US market or 2) whether the “border adjustment” on exports is an impermissible export subsidy. Thus, the DBCFT’s consistency with WTO rules is important in terms of international trade relations, but it does not necessarily mean the DBCFT is good economic policy.
  • Finally, just because something looks WTO-inconsistent doesn’t actually mean that it is (or that a WTO Member will challenge it). All WTO Members’ laws are considered to be consistent with WTO rules until found otherwise in official dispute settlement. Many Members therefore adopt policies that could be problematic under WTO rules, particularly when they think that the policies won’t be challenged, or political or economic considerations are more important that WTO ones. The threat of WTO retaliation and the appearance of being a responsible WTO Member are important variables, but they’re certainly not the only (or even predominant) ones.

So with those three things out of the way, let’s get on with the analysis. In general, a border adjustable internal tax will likely be consistent with WTO rules where—

1) The tax itself is among the types for which border adjustments are permitted. Under the WTOSubsidies Agreement, “direct taxes” (like a corporate income tax) are ineligible for border adjustment, and any such adjustment on export sales would constitute a prohibited export subsidy. This rule was put to the test in the late 90s when the United States lost a series of WTO disputes over its corporate income tax exemptions for “foreign sales corporations” (FSC/ETI): WTO panels and the Appellate Body found that the US corporate income tax was a “direct tax,” that the FSC/ETI tax breaks were “subsidies” to the recipient US firms, and that such tax breaks were contingent upon exportation—precisely the type of prohibited border tax adjustment subsidy listed in the Subsidies Agreement’s “Illustrative List” of prohibited subsidies.

On the other hand, the same WTO subsidy rules permit border adjustments on “indirect taxes” like value-added taxes (VATs), including the one applied by the EU. As a result, the conventional wisdom holds that only traditional consumption taxes (e.g., sales taxes or “credit-invoice” VATs that are paid by consumers and applied directly on products) are eligible for border adjustment under WTO rules. Those taxes certainly are eligible, but other taxes might also qualify. Most importantly, the definitions of “direct tax” and “indirect tax” in the WTO Subsidies Agreement leave possible gray areas for corporate taxes that share characteristics of both forms of taxation. This could include a “subtraction-method VAT,” under which a uniform rate of tax is levied directly on corporate sellers (as opposed to products/consumers) based on their sales revenue, less taxable (domestic) purchases.  Such a system is widely accepted as VAT (thus fitting the Subsidies Agreement’s definition of “indirect tax”), but it is also a tax on a form of corporate income (thus potentially meeting the “direct tax” definition). This ambiguity could protect a border adjustable, subtraction-method VAT from the relatively straightforward analysis that applied to the FSC/ETI measures. Indeed, Japan has imposed a border adjustable, subtraction-method VAT since the early 1990s, without any serious interest or concern from other WTO Members—a decent indication that the system doesn’t raise the same WTO alarm bells as FSC/ETI did. (Other WTO Members, of course, might disagree.)

2) The tax imposes identical burdens on imported goods and domestically produced “like” products. The General Agreement on Tariffs and Trade (GATT) generally prohibits taxes on imports other than duties, but Article II:2(a) allows a government to impose at the time a product crosses its border “a charge equivalent to an internal tax imposed…on a like domestic product,” as long as it is imposed consistently with the “national treatment” principle of GATT Article III. Thus, an internal tax at the border would be consistent with GATT Article II (on tariffs and import charges) and Article III (National Treatment) the imported good at issue is “not subject, directly or indirectly to internal taxes or other internal charges of any kind in excess of those applied directly or indirectly to like domestic products.”

3) The border adjustment on exports is no greater than the actual amount of tax collected or due. Under WTO rules, the rebate or exemption of eligible taxes (such as VAT) on exports will not be treated as an export subsidy, as long as the rebate/exemption rate is not greater than the rate at which the tax is levied domestically. On the other hand, such a rebate/exemption will constitute a prohibited export subsidy where it is in excess of the actual tax collected or due.

Based on these general rules, the DBCFT could, depending on its design, arguably pass muster at the WTO. First, in order for the tax exemption or rebate on US exports to avoid being designated a prohibited export subsidy, the adjusted corporate tax must not be a “direct tax,” as defined in the WTO Subsidies Agreement. The Republican proposal might pass this test if it is, for example, a subtraction method VAT. Second, the DBCFT must also be levied on imported products at a rate or amount no higher than the rate/amount levied on domestically produced “like” products (to be consistent with GATT Articles II and III). Third, the DBCFT cannot provide a border adjustment on export that is greater than the amount of tax actually levied or due (to avoid being a prohibited export subsidy).

All of these tests would require far more detail than we have at this stage, but the second issue may be the most problematic for the DBCFT. Various commenters have indicated that the DBCFT will permit certain additional deductions (e.g., for domestic wages and salaries) from the tax base, thus leading to a lower effective tax rate on domestic products than on the exact same imports. If that is, in fact, the case, then the DBCFT probably discriminates against imports in violation of GATT rules. We can test this once the legislative text is released through a simple hypothetical assessment of the tax’s effect on two identical U.S. companies selling and exporting the same product, with one company selling only imported final goods and the other selling identical products with 100% US content. If the total DBCFT paid by the former company is more than that paid by the latter (accounting for the whole value chain), then the tax measure would likely discriminate against imported goods in violation of GATT Articles II and III.  f the total export border adjustment provided to one of the companies is more than the tax collected (or otherwise due), then the system would likely be a prohibited export subsidy.

Obviously, the devil will be in the details. At this stage, I’m pretty agnostic about the DBCFT, but that certainly could change depending on the text of the final proposal and analyses of its projected economic (trade, budget, etc.) effects. In the meantime, however, we can dispense with the conventional wisdom that the DBCFT is definitely “protectionist” or that it definitely violates the United States’ WTO obligations. WTO-consistency, of course, doesn’t necessarily make the DBCFT good policy overall, but at least it would let us avoid a mess at the WTO and potential retaliation from our trading partners.