In light of all of the recent talk about the U.S. adopting a VAT-ish tax reform, I thought it was worth revisiting a fundamental question. Everyone seems to be operating under the assumption that a VAT is covered by GATT Article II:2(a) and Article III, and therefore a tax adjustment related to the VAT -- that is, a tax imposed on imports that is equivalent to the VAT imposed on domestic companies -- is permitted. But is it correct to say that a VAT is a tax on products that is covered by these provisions? A 1970 GATT working party report seemed to think so (para. 14), but today's world of GATT/WTO legal interpretation is very different. What would today's intensively textual analysis of Article II:2(a) and Article III look like on the issue of whether a VAT is a tax on products?
But before we get to that, there is the question of what exactly a VAT is. I'm sure there are lots of explanations, but I'll go with the one Joel recently pointed us to (another is here):
The example I always use is a loaf of bread you buy in a store for a buck -- so you have a farmer, a baker, and a supermarket along the production chain. Let's put the VAT at 10 percent.
1) The farmer grows the wheat and sells it to the baker for 20 cents. The VAT is 2 cents. The baker pays the farmer 22 cents, and the farmer sends 2 cents in VAT to the government.
2) The baker makes a loaf and sells it to the supermarket for 60 cents. The VAT is 6 cents. Now the supermarket pays the baker 66 cents, of which 6 is VAT. The baker sends the government 4 cents -- he pays 6 cents in VAT but receives a two cent credit from the government.3) The store sells the loaf to me for a dollar. I pay $1.10. The store sends the government 4 cents total - the 10 cents it collected in VAT on its sales, minus the 6 cents it paid to the baker in VAT, which it gets back in a credit. In total, the government gets 2 cents from the farmer, 4 cents from baker, 4 cents from the store. That's 10 cents on a final sale of a dollar -- for a 10 percent VAT.
This kind of VAT is referred to as a "credit-invoice method" VAT. There is also the "subtraction method" VAT, which I'll get to later.
Moving to the legal analysis, let's look at a key GATT provision. Tax adjustments related to imports would presumably violate GATT Article II:1, if they were not inscribed in the Schedule. But GATT Article II:2(a) says this:
2. Nothing in this Article shall prevent any contracting party from imposing at any time on the importation of any product:
(a) a charge equivalent to an internal tax imposed consistently with the provisions of paragraph 2 of Article III* in respect of the like domestic product ... ;
Thus, in order for a measure that would otherwise violate Article II to be justified under this provision, the charge on imports must be "equivalent to an internal tax ... in respect of the like domestic product." Along the same lines, a note to Article III states:
Any internal tax or other internal charge, or any law, regulation or requirement of the kind referred to in paragraph 1 which applies to an imported product and to the like domestic product and is collected or enforced in the case of the imported product at the time or point of importation, is nevertheless to be regarded as an internal tax or other internal charge, or a law, regulation or requirement of the kind referred to in paragraph 1, and is accordingly subject to the provisions of Article III.
This provision seems to take certain taxes and charges imposed on imported products out of Article II, and into Article III. And finally, Article III:2 says:
The products of the territory of any contracting party imported into the territory of any other contracting party shall not be 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. Moreover, no contracting party shall otherwise apply internal taxes or other internal charges to imported or domestic products in a manner contrary to the principles set forth in paragraph 1.
Now let's look at how these provisions could be applied to various types of taxes.
An excise tax, which applies to a particular good, would almost certainly be covered, and Article II:2(a)/Article III seem particularly well-suited for this kind of tax. Here, a specific, identifiable product can be identified. For example, if governments imposes excise duties on gasoline or cigarettes or alcohol, they will want to make sure they tax imports of those products as well as domestic products. If imposed at retail level, all producers are subject to the tax in the same way. It may also be the case that for some products, the tax gets paid at the manufacturing stage, whereas foreign products pay upon importation.
As we move on to other types of taxes, the connection to a product becomes more tenuous. A retail sales tax applies to a whole range of products, and is collected at the time of sale of the product. It is about products in general, or their sale at least, but not necessarily about specific products.
And then a VAT moves even further away from products, general or specific. The tax collection is not really about the product at all. Rather, the tax takes into account the "value added" along the way (the GAO says that "[a] value-added tax is a tax collected on the difference between a business’ sales and its purchases, otherwise known as the business’ 'value added.'"). It can be paid along the way as well. Thus, the tax is perhaps more about adding value at each stage of the production process than it is about the product. It is certainly not about taxing the product in the same way that, say, a gasoline tax is. And if that's the case, maybe it could be argued that, based on the text of Article II:2(a) and the language of Article III, a VAT is not covered by these provisions because it is not a tax on products at all. As a result, it would not be appropriate to impose a charge on imports that is equivalent to the VAT.
Now, perhaps this stretches the boundaries of interpretation a bit, and pushes too far away from the accepted GATT/WTO understanding of these issues. On the other hand, take a look at the following passages from John Jackson's World Trade and the Law of GATT (pp. 297-298), published in 1969, when the tax adjustment issue was a hot one:
It is doubtful that at the time the GATT was drafted the ultimate consequences of Articles II and III, as applied to border tax adjustments, were understood. In comparison to the then existing tariffs or other restrictive measures, the border tax adjustments probably seemed relatively inconsequential. But over the decades, as tariffs came down and quotas on manufactures were eliminated, other protection devices became more prominent. The border tax adjustment took on new significance and finally has emerged as a headline item and an issue of political attention. The reason for this is that countries that rely primarily on income taxes for revenues cannot, under the interpretations mentioned above, utilize border levies on imports to equalize the tax burden between imports and domestic products. On the other hand, countries that use a turnover tax system to collect the bulk of their revenues, use such a levy at their borders to impose an equal burden on imports, arguing that the turnover tax is a tax on "products" and within the meaning of Article III, paragraph 2. [FN 19] Thus, because of different tax structures and their relation to GATT, one country appears to have a considerable trade protection advantage over another.
-----------
FN 19. ... An argument can be made, however, that the turnover taxes do not qualify for the GATT border tax adjustment. This tax can be characterized roughly as follows: a tax on the gross receipts from the sale of a product, after subtracting from those gross receipts the costs of the materials purchased to make that product. It can be argued that this is not a tax on products at all, but a tax on income, or a certain type of income calculation based on earnings of the firm. Particularly, if it can be shown that the turnover tax reaches overhead of the firm, profits of the firm, management costs of the firm, transportation costs, such as gasoline for the trucks of the firm, etc., it can be argued in consequence that the turnover tax is not a tax on "products" within the GATT meaning.
Another approach is to recognize that the turnover tax is not the equivalent of an income tax such as that the United States imposes on firms, but to recognize also that tax schemes run the gamut from taxes directly on finished products, such as sales taxes or excise taxes at one end of the spectrum, to the other end of the spectrum where you have a tax on profits of the firm, apparently not related to the products produced. It can be argued that the turnover tax falls closer to the latter end of the spectrum than the former. In any event it can be argued that the turnover tax falls outside the category of "taxes on products" as that concept is used in GATT.
If this argument prevails, of course, then countries are not entitled under GATT to use border tax adjustments to add a charge on imports equivalent to domestic turnover taxes, any more than a country whose tax system depends primarily on income taxes would be entitled to use border tax adjustments to add a charge on imports equivalent to the domestic income tax burden that indirectly falls on the product.
One interesting aspect of his analysis is his description of the turnover tax at the beginning of the footnote, which sounds a lot like this description of the "subtraction method" VAT (much less common than the "credit-invoice method"):
Under [the subtraction method] VAT, the tax base is computed as the difference between the business’s taxable sales and its purchases of taxable goods and services. At the end of the reporting period, the tax rate is applied to the difference. The subtraction method VAT is considered to be “account-based,” as it is computed from the business’s books and records of account.
The way I read all this, Jackson was suggesting the argument -- not that he thought it was a winning argument, but he thought it credible enough to put out there -- that a particular kind of tax (the turnover tax), which happens to looks a lot like a kind of VAT, was not a tax imposed on a product. Therefore, this tax would not qualify for a tax adjustment.
Of course, the only reason any of this matters is because of the amounts involved. If all countries impose equivalent charges for imports in relation to a few excise duty products, or in relation to the fairly low retail sales tax, there will not be much concern about the trade impact. But when some countries collect a lot of tax revenue with a VAT (for which an adjustment has been available), whereas others use an income tax to generate most revenue (with no adjustment available), concerns about fairness are going to arise. There are certainly no easy answers here. But with the recent initiatives and talk coming from House Republicans and some Trump advisers, we may be forced to confront this issue again.
