III. Double Taxation Agreements

17.96. Australia has entered into double taxation agreements with a number of countries. The first of these was with the United Kingdom in 1946. Agreements with the United States, Canada and New Zealand were signed in 1953, 1957 and 1960 respectively. The original agreement with the United Kingdom was negotiated in 1968. The new agreement with that country owes much to the model OECD convention issued in 1963. This is true also of the agreements with Japan and Singapore signed in 1969, and of the one renegotiated with New Zealand in 1966. An agreement with West Germany was signed in 1972 but has yet to come into force. Australia has also entered into limited agreements with France and Italy in relation to airline profits.

17.97. Double taxation agreements reflect the revenue interests of the parties, their economic and social policies and, of course, their respective bargaining strength. They also reflect the concern of the parties to prevent injustice and discouragement of trade, investment and other contact between their residents which tend to result when the same income is subject to unrelieved double taxation.

17.98. The Committee does not propose to examine the compromises reached in particular treaties. It is concerned, however, to make some general observations on the structures and techniques of agreements that will be most effective in preventing double taxation. This of course is not to imply that the failure of an agreement to adopt these structures and techniques is a matter of inadequate expertise. Thus the failure to define the source of some kind of income may simply reflect the inability of the parties to agree to a compromise of revenue interests. The failure to define cloaks the problem of double taxation, it does not resolve it.

17.99. Those agreements subsequent to the OECD model convention of 1963 are distinctly sounder in structure and technique than earlier ones. The observations which follow, for the most part, express objectives that are already reflected in the OECD draft.

17.100. A double taxation agreement should resolve the conflicts of claims to tax that arise when a taxpayer is, by the law of each of the participating countries, resident in that country. The method of resolving the conflict involves adopting a notion of residence which will ensure that a person who is a dual resident has, for purposes of the agreement, only one residence. The country of residence, in the agreement sense, should be given the sole jurisdiction to tax the person's income from

  ― 271 ―
sources outside both countries. The other country should be denied jurisdiction to tax the person's income from sources in the country of residence.

17.101. These limitations on jurisdiction should also apply where there is no dual residence. The person may be resident by the law of only one country, which is thus the country of residence for purposes of the agreement. The jurisdiction of the other country should be limited so that it may not tax income having a source in a third country or in the country of residence. The United States claims to tax certain income of a foreign resident, even though it does not have a source in the United States, if that income is effectively connected with a trade or business which the foreign resident carries on in the United States. There is no express provision of the present double taxation agreement with the United States to prevent the United States from exercising this jurisdiction. Curiously, the United Kingdom agreement denies such jurisdiction to the country that is not the country of residence for purposes of the agreement in a dual residence situation, but not otherwise.

17.102. A double taxation agreement should define the meaning of source in relation to different kinds of income in ways which will ensure that the same income cannot be held to have a source in both countries. The definitions should apply to determine the meaning of source when jurisdiction to tax depends on source, and also in relation to the obligation of the country of residence to give credit for tax imposed in the country of source. The tendency in earlier agreements was to leave definitions of source to the operation of a provision in the agreement that ‘any term not otherwise defined shall, unless the context otherwise requires, have the meaning which it has under the law of the country applying the agreement’. There are wide divergencies of meaning of source in relation to different kinds of income in the laws of different countries. Dual source situations and unresolved double taxation problems must result if the matter is simply left to the law of the country applying the agreement. Moreover, the prospect is raised that the country applying the agreement may, in effect, rewrite the agreement unilaterally by changing a definition of source in its own law. There may be some justification for saying that the enactment by Australia of section 6C, defining the source of royalties, was a unilateral rewriting of the double taxation agreement with the United States.

17.103. The United Kingdom agreement, in relation to the obligation of the country of residence to give credit for tax imposed in the country of source requires, in some instances, that the country of residence accept the meaning of source given by the law of the country of source. This is some advance on an agreement leaving the matter to the country applying the agreement, but it is nonetheless unsatisfactory. The agreement should include its own definitions and not definitions imported by reference.

17.104. A double taxation agreement fixes the limits of each country's jurisdiction to tax in relation to different kinds of income. Here too it is of great importance that the agreement should contain its own definitions of terms. The United States agreement uses the term ‘royalties’ without its own definition of the term. The question has been raised whether Australia effectively extended the jurisdiction to tax given it by the agreement when, in 1968, it inserted a very wide definition of ‘royalties’ in the Act.