Empirical Test of Factor-Endowment Theory International Trade by Leontief Paradox

Many economists have tried to test the validity of Ohlin’s factor-endowment theorem with empirical findings. We shall review a few of them.

1. MacDougall made a pioneering attempt to find whether, a country’s export consists of goods involving relatively intensive use of its abundant factor or not. For this purpose, he compared England’s and America’s export shares of relatively capital-intensive goods and, to his surprise, he found that England did not export to America goods with low capital intensity relative to labour. That is to say, America, being a capital-abundant country, is found to be importing capital-intensive goods, which is quite contrary to the factor-endowment principle. As such, empirical finding seems to have refuted Ohlin’s theory.

However, this test of MacDougall has been severely criticised on the ground that his measure of capital intensity – in terms of horsepower used at prevailing factor prices – is not very satisfactory.

2. Then, Leontief (1956) measured and compared the factor intensities of American exports and imports by comparing the effects on the release of resources (Labour and Capital) under a reduction of one million dollars worth of exports if there is an equivalent import substitution. He, however, came to a paradoxical conclusion from his empirical study that, though, America is assumed to be capital-abundant country, her exports consist of labour-intensive goods and imports of capital- intensive products. In economic literature, this is commonly known as “Leontief Paradox.”

Leontief’s results were drawn by using the input-output analysis regarding U.S. data relating to 200 groups of industries, which were consolidated into 50 sectors, of which 38 traded their products directly on the international market. For this group of 38 sectors he computed the direct and indirect capital and labour requirements by using econometric methods of solving the simultaneous set of input-output equations. In fact, he computed capital and labour requirements for one million dollars worth of exports whose percentage composition was the same as total U.S. exports for 1947, and for U.S. production of one million dollars worth of import replacements. He obtained the following results:

Domestic Capital and Labour requirements per million dollars of exports and competitive import replacements (of average 1947 composition)

These computations reveal that in the United States in 1947, producers of import-competing goods employed 130 per cent more of capital than producers of export goods, with only 93 per cent of labour. That means U.S. exports were labour-intensive in 1947. Thus, America’s participation in the international division of labour is based on its specialisation on labour-intensive, rather than capital-intensive, lines of production. This finding is obviously contrary to Ohlin’s theory, because America is a capital-abundant country, yet its specialisation is found to be in labour-intensive lines of production. Tatemato and Ichimuara (1959) empiricals observed that Japan exported capital intensive products to the U.S. and other nations when in 1950s Japan was regarded as labour-abundant country.

The Leontief paradox is, however, not accepted as a fool-proof method of refuting Ohlin’s theorem. Leontief’s study has been criticised on the methodological ground that it was basically concerned with export industries and competitive import replacements rather than actual imports. Since, Ohlin’s theorem is concerned with actual exports and imports, it is obvious that Leontief’s conclusion is inapplicable to Ohlin’s theory and s cannot disprove it.

There are at least two possible explanations of the Leontief Paradox: (i) factor intensity reversal, and (ii) demand reversal. The presence of a factor intensity reversal nullifies the Heckscher-Ohlin theorem. But, empirical testing for a factor reversal on the basis of existing technological data is not an easy job. It is also difficult to say that, there is strong empirical evidence proving that the Leontief Paradox is the outcome of factor intensity reversal.

A demand reversal is perhaps a potential explanation of the Leontief Paradox. When a capital abundant country, for instance, is having a rising or very high domestic demand for capital-intensive product, it may have to import more of such product, then it contradicts with Heckscher-Ohlin theorem. When import replacements of America are to be considered, it is quite logical that American import replacement production would be more capital-intensive than export production, since American production techniques are highly capital-intensive; therefore, a considerable amount of America’s raw material imports would necessitate, relatively more capital if she were to produce them domestically rather than importing them. As such, she may find it profitable to import such capital- intensive goods rather than produce them at home, which would require relatively more of its abundant factor (capital).

R. Jones argues that, possibly both U.S. export and import competing goods might be produced by more capital-intensive techniques than those obtaining in other foreign countries. And since Ohlin’s theorem has no reversibility, one cannot say from trade patterns as to what relative factor endowments are. While Leontief did not make any effort to measure or compare relative factor endowments of America with those of other trading nations, he considered America only in a lopsided manner.

Further, Ohlin’s theory has been defended on the ground that perhaps foreign labour is less efficient than American labour, the use of capital per worker may be relatively smaller than that of many other nations. This means that U.S. is relatively labour-abundant (in a relative sense), so that, her exports would be of labour-intensive goods, which complies with the Ohlin theorem.

It has been also said that perhaps home demand for capital-intensive products in U.S. may be of very high and domestic supply may not be adequate to meet this demand, as such she is forced to import capital-intensive goods from abroad.

Hoffmeyer, however, argues that the Leontief Paradox is on account of the fact that he did not deal adequately with the natural resource component of goods. When goods necessitating a large proportion of natural resources are excluded from the estimate of capital-labour ratios, then Leontief’s conclusion would be just reversed. The U.S. exports will be found to be those of capital- intensive goods and imports will constitute labour-intensive products. Because, actual American imports have a significant natural resource element, and when Leontief was dealing with import replacements, it would be better to exclude natural resources commodities from the calculations. Besides, H.O. thereom assumed identical production functions. Whereas, Leontief used inappropriate methodology by assuming differential production functions for testing H.O. theory.

On the pattern of Leontief, Bharadwaj (1962) has made an empirical study of India’s bilateral trade with U.S. He found that Indian exports consist of capital-intensive goods to U.S. and her imports from U.S. of labour-intensive goods, which seems to be quite opposite of Ohlin’s doctrine. In this regard various explanations have been offered: typical climatic conditions, existence of a higher degree of disguised unemployment in Indian primary industries than in Indian manufacturing industries, extreme disparities in production technologies, etc.

Thus, these selected empirical tests appear to refute Ohlin’s theorem. But similar studies, such as those of Tatemoto and Ichimura of Japan’s bilateral trade, do not entirely refute Ohlin’s theory.

It has been, however, remarked that Ohlin’s theory has not really been tested so far. Further, Ohlin’s theory for all practical purposes is irrefutable, because it cannot be put to perfect empirical test on account of its unrealistic and restrictive assumptions – perfect competition and full employment.

In short, the H.O. theorem is plausible but cannot explain all trade: trade in all goods or the pattern of trade between all countries of the world economy. It lacks general validity since it is based on restricted assumptions.