In one of the most widely discussed tests of the factor proportions theory, Leontief attempted to reveal the relative factor proportions structure of U.S. participation in international trade.

It was considered that a country will tend to export those commodities which use its abundant factors of production intensively and import those which use its scarce factor intensively.

By common consent the United States is the only country that is most abundantly endowed with capital. Therefore, one would expected the United States to export capital intensive goods and import labour intensive goods.

Leontief’s first study was based on computation from input output tables constructed for the year 1947. Me computed for various industries the direct and indirect capital and labour required to produce a given dollar value of output. He then calculated the effects on capital and labour use of a given reduction in both U.S. imports and exports so that the relative commodity composition of exports and imports remained the same.


The conclusion was that the given value of U.S. exports embodied less capital and more labour than would be required to expand domestic output to provide an equivalent amount of competitive imports. Expressed inversely, U.S. import replacement industries required more capital relative to labour than did U.S. export industries.

The Leontief conclusion that in the international division of labour, the U.S. specialized in labour intensive rather than capital intensive goods contradicted the widely accepted view derived from the H.O. theory. Since it was not doubted that the U.S. was relatively capital abundant and relatively labour deficient, it would seem that, following the theory, exports should be capital intensive and import labour intensive. At first, there was no dispute over the H.O. proposition rather there was dispute over the particular empirical contradiction presented by Leontief.

How are we to explain Leontief’s paradoxical results that the most capital rich of all countries, the U.S. exports labour intensive goods?

Leontief himself explained the contradiction by reference to measures of labour supply. A concept more relevant that treating labour as a homogeneous item internationally and measuring it in man years would be treating it as “efficiency units” on which the U.S. because it has more productive labour, has relatively more efficiency units than it has units of a capital. Even working with the same amount of capital, the U.S. worker is more efficient than his foreign counterpart.


Leontief tried to explain his findings along two different lines. The one he gave priority ran in terms of differences in labour productivity. Leontief argued that American labour could not really be compared to labour in other countries, because the productivity of an American worker is substantially higher (three times higher, suggested Leontief) than that of foreign workers.

This would be one way, according to Leontief, by which his findings could be reconciled with the 11.0 theorem. Most economists might acknowledge the superior quality of U.S. labour. Leontief quotes a study by LB. Kravis indicating that wages are higher in U.S. export industries than in its import competing industries as supporting evidence. This however, conflicts with Leontief’s assumption of labour being a homogeneous factor of production, which would imply the same wage irrespective of occupations.

Another explanation for which Leontief has shown a certain understanding is connected with the two factor framework and the broad use of the term capital. The only two factors explicitly taken into account are labour and capital.

But as Leontief notes : ‘Invisible in all these tables but ever present as third factor or rather as whole additional set of factors determining this country’s productive capacity and, in particular, its comparative advantage vis-a-vis the rest of the world, are natural resources, agricultural lands, forests rivers and other rich mineral deposits.”


By taking into account this third factors an explanation to the Leontief’s paradox can be found. It might be the case, for instance, that imports require more capital to labour than exports; it is still, however, possible that imports are intensive in the third factor, say land. If capital and the third factor (land) are substitutes but both are complementary with labour, it might be the case that import competing goods are capital intensive in the U.S. but land intensive abroad. By bringing a third factor, in to account in this way, possible explanation might be found.

In his analysis, Leontief took only one country into account only computed factor requirements for marginal changes in the production of American exports and import competing goods. If factor reversals exist, it is fully possible for a capital rich country to export its labour intensive goods. The country will still use more capital intensive methods in its export industries than any other country. Leontief never brought a second country into account. Had he done so and compared, for instance, the factor intensities in American export industries with those of Japan or Western Europe, he might well have found that American exports were capital intensive compared to Japan or Western Europe exports. According to R.W. Jones, by invoking factor reversals we can thus explain Leontief’s puzzling results.

Another explanation to the Leontief’s paradox has been given by Erik Hoff Meyer. He argues that if products relying to a large extent on natural resources are excluded from Leontief’s list of goods, the normally expected picture that the U.S. exports capital intensive goods and imports labour intensive goods will prevail.

W.P. Travis explains the Leontief paradox with the help of U.S. trade policy. He refers to the fact that U.S. trade is highly protected, a fact even more true when Leontief made his study than it is today. When Leontief made his study, most competitive imports considered of crude oil, paper pulp, primary copper and lead, and metallic ores. These products were imported because the U.S. simply could not produce them. These products are more capital intensive than any other products. According to Travis, U.S. protective trade policies alone, therefore, are sufficient to explain the Leontief paradox.

The discussion of the Leontief paradox has hardly been able to establish firm conclusions. It has provided a good deal of insight into the foreign trade position of the U.S., but it has hardly helped to establish or refute the H.O. theory of international trade. Static theories such as the factor price equalization theorem or the H.O. theory of comparative advantage should perhaps not primarily be viewed as geared toward empirical testing. They are first and foremost means of studying the general equilibrium characteristics of open economies.

By studying the H.O. trade model, we have been able to get a good understanding of the meaning of general equilibrium. We have seen how the possibility of trade causes a change in commodity prices, giving rise to a change in factor prices, to a reallocation of factors of production, and a change in the production structure. All these variables are intimately linked together and it is not possible to change one of them without changing all the others.