41. Memorandum From Secretary of the Treasury Kennedy1


  • The Honorable Maurice Stans
  • The Honorable George Shultz
  • The Honorable Robert Mayo
  • The Honorable Paul W. McCracken
  • The Honorable Nathaniel Samuels
  • The Honorable Carl Gilbert


  • Tax Adjustments at the Border

Chairman McCracken’s memorandum of June 8 suggests that we proceed with the three points outlined by Ambassador Gilbert while setting aside the issue of the basic inequity of GATT rules.2 I do not believe that approach would relieve either our economic or political difficulties. Of the three points proposed by Ambassador Gilbert, two are clearly of minimal importance and solutions would result in no substantial trade benefits for U.S. producers. The third faces us with the same basic issue of GATT inequity which Chairman McCracken suggests we set aside. Any substantive proposal on changes would require an amendment of GATT provisions concerning the amount of allowable adjustments.

[Page 103]

The U.S. has talked about taxe occulte3 and averaging primarily for tactical purposes—keeping the talks alive while we consider the basic issue. The problems of averaging and border adjustments for taxe occulte have largely passed us by as they do not inherently exist in the value added tax system. As Italy and Belgium will be adopting the TVA within 18 months, only Spain and Austria, among the developed nations, will be left with cascade tax systems4—the area of most abuse regarding averaging and taxe occulte. A modification of taxe occulte procedures would limit possible U.S. action while leaving Europeans free to obtain benefits equivalent to adjustment for taxe occulte by simple modifications of their TVA systems. It is clear that there is little economic or political advantage in pursuing a change regarding these points.

As for the third point, I agree that countries should not be allowed unilaterally to disrupt the international trading mechanism by changes in their border adjustments.

Border tax adjustments will continue to be a problem as EC tax harmonization proceeds. Eventually all of Western Europe will be using the TVA and making substantial changes in their border adjustments. These changes, condoned by the bias in the GATT rules, will have serious disruptive effects on both trade and international balance of payments adjustments. Failure to resist this undercutting of our economic strength will badly damage our ability to prevent other similar actions.

In order to argue that changes should be controlled, we must demonstrate that they have trade effects. But in most instances this is true only if direct taxes are, in part or in whole, passed forward to the consumer and/or indirect taxes are partially absorbed by the producer. Either position directly contradicts GATT rules and confronts us with the issue of amending them to correct the bias in favor of indirect tax systems. Unless the rules are amended, countries would argue that their actions are in conformity with GATT and they have no responsibility to offset any trade effects of changes in adjustments.

Thus advocacy by the U.S. of proposals covering the points raised by Ambassador Gilbert would seem to make sense only as part of a [Page 104]package which includes a major change in how nations handle border adjustments for taxes.

Chairman McCracken’s thesis that past changes in tax adjustments at the border are washed out by exchange rate changes disturbs me.5 It seems to me wrong in implying an equilibrium that simply does not exist and cannot practicably be obtained.

We have all recognized the absolute necessity of attaining a stronger goods and services position. The present bias in the border tax adjustment rules complicates the achievement of this goal.

The plain fact is that exchange rate changes of the last 10 or 15 years have not and will not eliminate the problem of existing border tax adjustments: our trade balance and balance of payments structure have deteriorated in recent years. The fact that some exchange rate changes might have been different without the border adjustments, if true, provides no answer to the U.S. structural problem. Furthermore, numerous changes in border adjustments have occurred which were not offset even partially by exchange adjustments. Thus, Belgium and Italy have not modified their exchange rates since 1949, the Dutch since 1961, and most of Scandinavia since the immediate post World War II period. Changes in taxes and border adjustments have occurred regularly, with rates and product coverage generally increasing. It is only with respect to the 1960 and 1968 German revaluation and the 1961 Dutch revaluation that we can conceivably say that exchange rate change even went in the right direction in order to offset in part the trade effects of the border adjustment. But even in those cases, it cannot be definitively stated that the trade effects of cumulative border tax adjustments were effectively offset. It seems to me fruitless to argue that remaining disequilibria can simply be offset by further exchange rate changes that in practice are both unlikely in the degree necessary and deeply disturbing to the international monetary climate.

Carried to its logical conclusion, Chairman McCracken’s argument implies that the U.S. need not worry about the level of existing U.S. and foreign tariffs, U.S. and foreign subsidies or most U.S. and foreign import barriers as changes in exchange rates have eliminated their economic impact on U.S. and foreign trade interests. If this were so, the trade message submitted by the President need not have called for tariff reducing authority nor provided for retaliatory authority against foreign subsidies in third country markets. Although exchange rate [Page 105]changes may conceivably eliminate balance of payments disequilibrium, in the sense of reserve losses and gains, we must always question whether the process of adjustment is desirable, the new equilibrium is appropriate for the world and for the U.S., and the resulting payment structure and resource allocation are truly efficient. A new equilibrium with the EC in a heavy trade surplus and the U.S. relying on capital inflows would be structurally unsatisfactory for the U.S. and for the entire world.

On a political level, I also do not believe that an argument that exchange adjustments have eliminated the impact of old border adjustments will be persuasive. Certainly these exchange adjustments do not eliminate our countervailing duty problems as the border adjustments continue to exist. In this regard I would point to recent statements by Congressman Mills that he intends to amend the countervailing duty law to require action against all rebates of taxes.

As I mentioned before, any effective mechanism for controlling changes in border adjustments must have as its basis the same arguments already put forward on the amount of adjustment for direct and indirect taxes. To achieve an effective control limiting a country’s ability to make such adjustments or changes in them would require a basic amendment to the GATT rules. By limiting our substantive proposals to controlling changes in adjustments we do not reduce the need for achieving a structural change in GATT. We would, however, have thrown out one of our basic arguments, receiving nothing in exchange, and prejudicing our credibility on other U.S. initiatives.

David M. Kennedy 6
  1. Source: Washington National Records Center, Department of the Treasury, Secretary’s Memos/Correspondence: FRC 56 74 7, Council of Economic Advisers. Confidential. Forwarded to Kennedy under cover of a June 25 memorandum from Petty.
  2. McCracken’s June 8 memorandum summarized the results of a June 5 meeting, where agreement was reached on how to proceed at the July GATT meeting. Gilbert outlined three points: “opposition to adjustments for taxes occultes; a requirement for confrontation and justification in the event of changes in a country’s tax system involving border adjustments, and international control or surveillance of ‘averaging.’” GATT rules allowed Contracting Parties to levy border taxes, sometimes known as border tax adjustments, imposing domestic, indirect taxes (i.e., excise, value added, and turnover taxes) on imports and rebating and/or excusing such taxes on their exports. A number of European nations and Japan, which relied heavily on indirect taxes (such as value added and turnover taxes), imposed significant border taxes, whereas the United States, which relied primarily on direct taxes (particularly income and property taxes), had only very limited scope for making border tax adjustments. In many circles this was viewed as discrimination against U.S. exports and subsidization of imports into the United States, contributing significantly to the U.S. balance-of-payments deficit. A number of papers regarding border taxes are in the Volcker Group records in the Washington National Records Center, Department of the Treasury, Volcker Group Masters: FRC 56 86 30. Border taxes were discussed in the GATT Working Party on Border Tax Adjustments. Documentation is in the National Archives, RG 59, Central Files 1970-72, FN 16 GATT.
  3. The taxe occulte is the “hidden” amount of tax that accrues in the value of a product, depending on the number of transactions that occur during a product’s production and distribution. Unlike value added taxes where the rate of application is generally clear, when taxe occulte occurs the effective rate is difficult to gauge, giving rise to the question of what is the appropriate, “average” rate for border tax purposes. See Border Tax Adjustments and Tax Structures in OECD Member Countries (Paris: Organization for Economic Cooperation and Development, 1968), pp. 20-21 and 58-63.
  4. The cascade tax, or the turnover tax, was used in several European countries. Community members were expected to replace their cascade taxes with value added taxes.
  5. McCracken argued that if a country made a 10 percent border tax adjustment for, say, a value added tax, by rebating that amount on exports and levying that amount on imports, any trade impact of that adjustment would be offset by a corresponding 10 percent appreciation in that country’s currency, which would render its exports 10 percent more expensive in foreign currencies and its imports 10 percent more expensive in domestic currency.
  6. Printed from a copy that indicates Kennedy signed the original.