Empirical Evidence on the Heckscher-Ohlin Model
The essence of the Heckscher-Ohlin model is that trade is driven by differences in factor
abundance across countries. We just saw how this leads to the natural prediction that goods
trade is substituting for factor trade, and hence that goods trade across countries should
embody those factor differences. This is a very powerful prediction that can be tested empirically.
However, we will see that the empirical successes of such tests are very limited—
mainly due to the same reasons that undermine the prediction for factor-price equalization
(especially the assumption of common technologies across countries). Does this mean that
differences in factor abundance do not help explain the observed patterns of trade across
countries? Not at all. We will see how the pattern of trade between developed and developing
countries does fit quite well with the predictions of the Heckscher-Ohlin model.
Trade in Goods as a Substitute for Trade in Factors
Tests on U.S. Data Until recently, and to some extent even now, the United States has
been a special case among countries. Until a few years ago, the United States was much
wealthier than other countries, and U.S. workers visibly worked with more capital per
person than their counterparts in other countries. Even now, although some Western
European countries and Japan have caught up, the United States continues to be high on
the scale of countries as ranked by capital-labor ratios.
One would then expect the United States to be an exporter of capital-intensive goods and
an importer of labor-intensive goods. Surprisingly, however, this was not the case in the
25 years after World War II. In a famous study published in 1953, economist Wassily Leontief
(winner of the Nobel Prize in 1973) found that U.S. exports were less capital-intensive than
U.S. imports.11 This result is known as the Leontief paradox.
11See Wassily Leontief, “Domestic Production and Foreign Trade: The American Capital Position Re-Examined,”
Proceedings of the American Philosophical Society 97 (September 1953), pp. 331–349.
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 99
Table 5-2 illustrates the Leontief paradox as well as other information about U.S. trade
patterns. We compare the factors of production used to produce $1 million worth of 1962
U.S. exports with those used to produce the same value of 1962 U.S. imports. As the first
two lines in the table show, Leontief’s paradox was still present in that year: U.S. exports
were produced with a lower ratio of capital to labor than U.S. imports. As the rest of the table
shows, however, other comparisons of imports and exports are more in line with what one
might expect. The United States exported products that were more skilled-labor-intensive
than its imports, as measured by average years of education. We also tended to export products
that were “technology-intensive,” requiring more scientists and engineers per unit of
sales. These observations are consistent with the position of the United States as a high-skill
country, with a comparative advantage in sophisticated products.
Why, then, do we observe the Leontief paradox? Some studies have argued that this
paradox was specific to the time period considered.12 Others point to the needed
assumption of common technologies used by the United States and its trading partners,
which is likely to be violated. One such violation that would explain the paradox goes
as follows: The United States has a special advantage in producing new products or
goods made with innovative technologies, such as aircraft and sophisticated computer
chips. Such products may well be less capital-intensive than products whose technology
has had time to mature and become suitable for mass production techniques. Thus
the United States may be exporting goods that heavily use skilled labor and innovative
entrepreneurship, while importing heavy manufactures (such as automobiles) that use
large amounts of capital.
Tests on Global Data Since the United States may be a special case, economists have
also attempted to broaden the test to incorporate more countries, as well as more factors of
production. An important such study by Harry P. Bowen, Edward E. Leamer, and Leo
Sveikauskas13 extended the predictions for the factor content of trade to 27 countries and
12 factors of production. The theory behind the test is the same as for Leontief’s test for
the United States: Based on the factor content of exports and imports, a country should be
a net exporter of a factor of production with which it is relatively abundantly endowed
(and conversely, net importer of those with which it is relatively poorly endowed).
12Later studies point to the disappearance of the Leontief paradox by the early 1970s. For example, see Robert
M. Stern and Keith E. Maskus, “Determinants of the Structure of U.S. Foreign Trade, 1958–76,” Journal of
International Economics 11 (May 1981), pp. 207–224. These studies show, however, the continuing importance
of human capital in explaining U.S. exports.
13See Harry P. Bowen, Edward E. Leamer, and Leo Sveikauskas, “Multicountry, Multifactor Tests of the Factor
Abundance Theory,” American Economic Review 77 (December 1987), pp. 791–809.
TABLE 5-2 Factor Content of U.S. Exports and Imports for 1962
Imports Exports
Capital per million dollars $2,132,000 $1,876,000
Labor (person-years) per million dollars 119 131
Capital-labor ratio (dollars per worker) $17,916 $14,321
Average years of education per worker 9.9 10.1
Proportion of engineers and scientists in work force 0.0189 0.0255
Source: Robert Baldwin, “Determinants of the Commodity Structure of U.S. Trade,” American Economic
Review 61 (March 1971), pp. 126–145.
100 PART ONE International Trade Theory
Table 5-3 shows one of the key tests of Bowen et al. The authors calculated the ratio of
each country’s endowment of each factor to the world supply of that factor. They then
compared these ratios with each country’s share of world income. If the factor-proportions
theory was right, a country would always export factors for which the factor share
exceeded the income share, and import factors for which it was less. In fact, for two-thirds
of the factors of production, trade ran in the predicted direction less than 70 percent of the
time. This result confirms the Leontief paradox on a broader level: Trade often does not
run in the direction that the Heckscher-Ohlin theory predicts. As with the Leontief paradox
for the United States, explanations for this result have centered on the failure of the
common technology assumption.
The Case of the Missing Trade Another indication of large technology differences
across countries comes from discrepancies between the observed volumes of trade and
those predicted by the Heckscher-Ohlin model. In an influential paper, Daniel Trefler14 at
the University of Toronto pointed out that the Heckscher-Ohlin model can also be used to
derive predictions for a country’s volume of trade based on differences in that country’s
factor abundance with that of the rest of the world (since, in this model, trade in goods is
substituting for trade in factors). In fact, factor trade turns out to be substantially smaller
than the Heckscher-Ohlin model predicts.
A large part of the reason for this disparity comes from a false prediction of largescale
trade in labor between rich and poor nations. Consider the United States, on one
side, and China on the other. In 2008, the United States had about 23 percent of world
income but only about 5 percent of the world’s workers; so a simple factor-proportions
theory would suggest that U.S. imports of labor embodied in trade should have been
huge, something like four times as large as the nation’s own labor force. In fact,
calculations of the factor content of U.S. trade showed only small net imports of labor.
Conversely, China had 7 percent of world income but approximately 20 percent of
TABLE 5-3 Testing the Heckscher-Ohlin Model
Factor of Production Predictive Success*
Capital 0.52
Labor 0.67
Professional workers 0.78
Managerial workers 0.22
Clerical workers 0.59
Sales workers 0.67
Service workers 0.67
Agricultural workers 0.63
Production workers 0.70
Arable land 0.70
Pasture land 0.52
Forest 0.70
*Fraction of countries for which net exports of factor runs in predicted direction.
Source: Harry P. Bowen, Edward E. Leamer, and Leo Sveikauskas, “Multicountry, Multifactor Tests of
the Factor Abundance Theory,” American Economic Review 77 (December 1987), pp. 791–809.
14Daniel Trefler, “The Case of the Missing Trade and Other Mysteries,” American Economic Review 85
(December 1995), pp. 1029–1046.
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 101
the world’s workers in 2008; it therefore “should” have exported most of its labor via
trade—but it did not.
Allowing for technology differences also helps to resolve this puzzle of “missing
trade.” The way this resolution works is roughly as follows: If workers in the United States
are much more efficient than those in China, then the “effective” labor supply in the
United States is much larger compared with that of China than the raw data suggest—and
hence the expected volume of trade between labor-abundant China and labor-scarce
America is correspondingly less.
If one makes the working assumption that technological differences between countries
take a simple multiplicative form—that is, that a given set of inputs produces only times
as much in China as it does in the United States, where is some number less than 1—it is
possible to use data on factor trade to estimate the relative efficiency of production in different
countries. Table 5-4 shows Trefler’s estimates for a sample of countries; they suggest
that technological differences are in fact very large. However, this exercise does not
prove that technology differences do have this simple multiplicative form. If they don’t,
then some country could have bigger technological advantages in particular sectors, and
the predictions for the pattern of trade would be a mix between those of the Ricardian and
Hecksher-Ohlin models.
Patterns of Exports Between Developed
and Developing Countries
Although the overall pattern of international trade does not seem to be very well accounted
for by a pure Heckscher-Ohlin model, comparisons of the exports of labor-abundant, skillscarce
nations in the third world with the exports of skill-abundant, labor-scarce nations do
fit the theory quite well. Consider, for example, Figure 5-12, which compares the pattern
of U.S. imports from Bangladesh, whose work force has low levels of education, with the
pattern of U.S. imports from Germany, which has a highly educated labor force.
In Figure 5-12, which comes from the work of John Romalis of the University of
Chicago,15 goods are ranked by skill intensity: the ratio of skilled to unskilled labor used
in their production. The vertical axes of the figure show U.S. imports of each good from
Germany and Bangladesh, respectively, as a share of total U.S. imports of that good. As
you can see, Bangladesh tends to account for a relatively large share of U.S. imports of
low-skill-intensity goods such as clothing, but a low share of highly skill-intensive goods.
Germany is in the reverse position.
d
d
TABLE 5-4 Estimated Technological Efficiency, 1983 (United States = 1)
Country
Bangladesh 0.03
Thailand 0.17
Hong Kong 0.40
Japan 0.70
West Germany 0.78
Source: Daniel Trefler, “The Care of the Missing Trade and Other Mysteries,” American Economic Review
85 (December 1995), pp. 1029–1046.
15John Romalis, “Factor Proportions and the Structure of Commodity Trade,” American Economic Review 94
(March 2004), pp. 67–97.
102 PART ONE International Trade Theory
Changes over time also follow the predictions of the Heckscher-Ohlin model. Figure 5-13
shows the changing pattern of exports to the United States from Western Europe, Japan, and
the four Asian “miracle” economies—South Korea, Taiwan, Hong Kong, and Singapore—
which moved rapidly from being quite poor economies in 1960 to relatively rich
economies with highly skilled work forces today.
Panel (a) of Figure 5-13 shows the pattern of exports from the three groups in 1960; the
miracle economies were clearly specialized in exports of low-skill-intensity goods, and
even Japan’s exports were somewhat tilted toward the low-skill end. As shown in panel
(b), by 1998, however, the level of education of Japan’s work force was comparable to that
of Western Europe, and Japan’s exports reflected that change, becoming as skill-intensive
as those of European economies. Meanwhile, the four miracle economies, which had rapidly
increased the skill levels of their own work forces, had moved to a trade pattern comparable
to that of Japan a few decades earlier.
A key prediction of the Heckscher-Ohlin model is that changes in factor abundance
lead to biased growth toward sectors that use that factor intensively in production. We can
see that the experience of those Asian economies fit very well with these predictions: As
the supply of skilled labor increased, they increasingly specialized in the production of
skill-intensive goods.
Implications of the Tests
We have just seen that the empirical testing of the Heckscher-Ohlin model has produced
mixed results. In particular, the evidence is weak concerning the prediction of the model
that, absent technology differences between countries, trade in goods is a substitute for
trade in factors: The factor content of a country’s exports does not always reflect that
0.06
0.08
0.10
0.12
Estimated share of US imports
by industry
Estimated share of US imports
by industry
0.004
0.003
0.002
0.001
0.000
0.04
0.02
Germany
(left scale)
Bangladesh
(right scale)
Skill intensity of industry
0.00
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Figure 5-12
Skill Intensity and the Pattern of U.S. Imports from Two Countries
Source: John Romalis, “Factor Proportions and the Structure of Commodity Trade,” American Economic Review 94
(March 2004), pp. 67–97.
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 103
country’s abundant factors; and the volume of trade is substantially lower than what would
be predicted based on the large differences in factor abundance between countries.
However, the pattern of goods trade between developed and developing countries fits the
predictions of the model quite well.
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Share of U.S. imports by industry
Four miracles
Japan
Western Europe
Skill intensity of industry
0.05
(a) 1960
(a) 1998
0.10 0.15
1960
1998
0.20 0.25 0.30 0.35 0.40
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Share of U.S. imports by industry
Four miracles
Japan
Western Europe
Skill intensity of industry
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Figure 5-13
Changing Patterns of Comparative Advantage
104 PART ONE International Trade Theory
The Heckscher-Ohlin model also remains vital for understanding the effects of trade,
especially its effects on the distribution of income. Indeed, the growth of North-South
trade in manufactures—a trade in which the factor intensity of the North’s imports is very
different from that of its exports—has brought the factor-proportions approach into the
center of practical debates over international trade policy.
SUMMARY
1. To understand the role of resources in trade, we develop a model in which two goods
are produced using two factors of production. The two goods differ in their factor
intensity, that is, at any given wage-rental ratio, production of one of the goods will use
a higher ratio of capital to labor than production of the other.
2. As long as a country produces both goods, there is a one-to-one relationship between the
relative prices of goods and the relative prices of factors used to produce the goods. A rise
in the relative price of the labor-intensive good will shift the distribution of income in
favor of labor, and will do so very strongly: The real wage of labor will rise in terms of
both goods, while the real income of capital owners will fall in terms of both goods.
3. An increase in the supply of one factor of production expands production possibilities,
but in a strongly biased way: At unchanged relative goods prices, the output of the
good intensive in that factor rises while the output of the other good actually falls.
4. A country that has a large supply of one resource relative to its supply of other
resources is abundant in that resource. A country will tend to produce relatively more
of goods that use its abundant resources intensively. The result is the basic Heckscher-
Ohlin theory of trade: Countries tend to export goods that are intensive in the factors
with which they are abundantly supplied.
5. Because changes in relative prices of goods have very strong effects on the relative
earnings of resources, and because trade changes relative prices, international trade
has strong income distribution effects. The owners of a country’s abundant factors gain
from trade, but the owners of scarce factors lose. In theory, however, there are still
gains from trade, in the limited sense that the winners could compensate the losers,
and everyone would be better off.
6. In an idealized model, international trade would actually lead to equalization of the
prices of factors such as labor and capital between countries. In reality, complete
factor-price equalization is not observed because of wide differences in resources, barriers
to trade, and international differences in technology.
7. Empirical evidence is mixed on the Heckscher-Ohlin model, but most researchers
do not believe that differences in resources alone can explain the pattern of world
trade or world factor prices. Instead, it seems to be necessary to allow for substantial
international differences in technology. Nonetheless, the Heckscher-Ohlin
model does a good job of predicting the pattern of trade between developed and
developing countries.
KEY TERMS
abundant factor, p. 91
biased expansion of production
possibilities, p. 88
equalization of factor
prices, p. 97
factor abundance, p. 80
factor intensity, p. 80
factor prices, p. 85
factor-proportions
theory, p. 80
Heckscher-Ohlin theory, p. 80
Leontief paradox, p. 98
scarce factor, p. 91
skill-biased technological
change, p. 95
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 105
PROBLEMS
1. Go back to the numerical example with no factor substitution that leads to the production
possibility frontier in Figure 5-1.
a. What is the range for the relative price of cloth such that the economy produces
both cloth and food? Which good is produced if the relative price is outside of this
range?
For parts (b) through (f), assume that the price range is such that both goods are
produced.
b. Write down the unit cost of producing one yard of cloth and one calorie of food as
a function of the price of one machine-hour, r, and one work-hour, w. In a competitive
market, those costs will be equal to the prices of cloth and food. Solve for the
factor prices r and w.
c. What happens to those factor prices when the price of cloth rises? Who gains and
who loses from this change in the price of cloth? Why? Do those changes conform
to the changes described for the case with factor substitution?
d. Now assume that the economy’s supply of machine-hours increases from 3,000 to
4,000. Derive the new production possibility frontier.
e. How much cloth and food will the economy produce after this increase in its
capital supply?
f. Describe how the allocation of machine-hours and work-hours between the cloth
and food sectors changes. Do those changes conform with the changes described
for the case with factor substitution?
2. In the United States, where land is cheap, the ratio of land to labor used in cattle
raising is higher than that of land used in wheat growing. But in more crowded
countries, where land is expensive and labor is cheap, it is common to raise cows
by using less land and more labor than Americans use to grow wheat. Can we still
say that raising cattle is land-intensive compared with farming wheat? Why or
why not?
3. “The world’s poorest countries cannot find anything to export. There is no resource
that is abundant—certainly not capital or land, and in small poor nations not even
labor is abundant.” Discuss.
4. The U.S. labor movement—which mostly represents blue-collar workers rather than
professionals and highly educated workers—has traditionally favored limits on
imports from less-affluent countries. Is this a shortsighted policy or a rational one in
view of the interests of union members? How does the answer depend on the model
of trade?
5. Recently, computer programmers in developing countries such as India have begun
doing work formerly done in the United States. This shift has undoubtedly led to
substantial pay cuts for some programmers in the United States. Answer the following
two questions: How is this possible, when the wages of skilled labor are
rising in the United States as a whole? What argument would trade economists
make against seeing these wage cuts as a reason to block outsourcing of computer
programming?
6. Explain why the Leontief paradox and the more recent Bowen, Leamer, and Sveikauskas
results reported in the text contradict the factor-proportions theory.
7. In the discussion of empirical results on the Heckscher-Ohlin model, we noted
that recent work suggests that the efficiency of factors of production seems to differ
internationally. Explain how this would affect the concept of factor-price
equalization.
106 PART ONE International Trade Theory
FURTHER READINGS
Donald R. Davis and David E. Weinstein. “An Account of Global Factor Trade.” American
Economic Review 91 (December 2001), pp. 1423–1453. The authors review the history of tests of
the Heckscher-Ohlin model and propose a modified version—backed by extensive statistical
analysis—that allows for technology differences, specialization, and transportation costs.
Alan Deardorff. “Testing Trade Theories and Predicting Trade Flows,” in Ronald W. Jones and Peter
B. Kenen, eds. Handbook of International Economics. Vol. 1. Amsterdam: North-Holland, 1984.
A survey of empirical evidence on trade theories, especially the factor-proportions theory.
Gordon Hanson and Ann Harrison. “Trade and Wage Inequality in Mexico.” Industrial and Labor
Relations Review 52 (1999), pp. 271–288. A careful study of the effects of trade on income
inequality in our nearest neighbor, showing that factor prices have moved in the opposite direction
from what one might have expected from a simple factor-proportions model. The authors
also put forward hypotheses about why this may have happened.
Ronald W. Jones. “Factor Proportions and the Heckscher-Ohlin Theorem.” Review of Economic
Studies 24 (1956), pp. 1–10. Extends Samuelson’s 1948–1949 analysis (cited below), which
focuses primarily on the relationship between trade and income distribution, into an overall
model of international trade.
Ronald W. Jones. “The Structure of Simple General Equilibrium Models.” Journal of Political
Economy 73 (December 1965), pp. 557–572. A restatement of the Heckscher-Ohlin-Samuelson
model in terms of elegant algebra.
Ronald W. Jones and J. Peter Neary. “The Positive Theory of International Trade,” in Ronald W.
Jones and Peter B. Kenen, eds. Handbook of International Economics. Vol. 1. Amsterdam:
North-Holland, 1984. An up-to-date survey of many trade theories, including the factorproportions
theory.
Bertil Ohlin. Interregional and International Trade. Cambridge: Harvard University Press, 1933.
The original Ohlin book presenting the factor-proportions view of trade remains interesting—its
complex and rich view of trade contrasts with the more rigorous and simplified mathematical
models that followed.
Robert Reich. The Work of Nations. New York: Basic Books, 1991. An influential tract that argues
that the increasing integration of the United States in the world economy is widening the gap
between skilled and unskilled workers.
John Romalis. “Factor Proportions and the Structure of Commodity Trade.” The American
Economic Review 94 (March 2004), pp. 67–97. A recent, state-of-the-art demonstration that a
modified version of the Heckscher-Ohlin model has a lot of explanatory power.
Paul Samuelson. “International Trade and the Equalisation of Factor Prices.” Economic Journal 58
(1948), pp. 163–184; and “International Factor Price Equalisation Once Again.” Economic
Journal 59 (1949), pp. 181–196. The most influential formalizer of Ohlin’s ideas is Paul
Samuelson (again!), whose two Economic Journal papers on the subject are classics.
The essence of the Heckscher-Ohlin model is that trade is driven by differences in factor
abundance across countries. We just saw how this leads to the natural prediction that goods
trade is substituting for factor trade, and hence that goods trade across countries should
embody those factor differences. This is a very powerful prediction that can be tested empirically.
However, we will see that the empirical successes of such tests are very limited—
mainly due to the same reasons that undermine the prediction for factor-price equalization
(especially the assumption of common technologies across countries). Does this mean that
differences in factor abundance do not help explain the observed patterns of trade across
countries? Not at all. We will see how the pattern of trade between developed and developing
countries does fit quite well with the predictions of the Heckscher-Ohlin model.
Trade in Goods as a Substitute for Trade in Factors
Tests on U.S. Data Until recently, and to some extent even now, the United States has
been a special case among countries. Until a few years ago, the United States was much
wealthier than other countries, and U.S. workers visibly worked with more capital per
person than their counterparts in other countries. Even now, although some Western
European countries and Japan have caught up, the United States continues to be high on
the scale of countries as ranked by capital-labor ratios.
One would then expect the United States to be an exporter of capital-intensive goods and
an importer of labor-intensive goods. Surprisingly, however, this was not the case in the
25 years after World War II. In a famous study published in 1953, economist Wassily Leontief
(winner of the Nobel Prize in 1973) found that U.S. exports were less capital-intensive than
U.S. imports.11 This result is known as the Leontief paradox.
11See Wassily Leontief, “Domestic Production and Foreign Trade: The American Capital Position Re-Examined,”
Proceedings of the American Philosophical Society 97 (September 1953), pp. 331–349.
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 99
Table 5-2 illustrates the Leontief paradox as well as other information about U.S. trade
patterns. We compare the factors of production used to produce $1 million worth of 1962
U.S. exports with those used to produce the same value of 1962 U.S. imports. As the first
two lines in the table show, Leontief’s paradox was still present in that year: U.S. exports
were produced with a lower ratio of capital to labor than U.S. imports. As the rest of the table
shows, however, other comparisons of imports and exports are more in line with what one
might expect. The United States exported products that were more skilled-labor-intensive
than its imports, as measured by average years of education. We also tended to export products
that were “technology-intensive,” requiring more scientists and engineers per unit of
sales. These observations are consistent with the position of the United States as a high-skill
country, with a comparative advantage in sophisticated products.
Why, then, do we observe the Leontief paradox? Some studies have argued that this
paradox was specific to the time period considered.12 Others point to the needed
assumption of common technologies used by the United States and its trading partners,
which is likely to be violated. One such violation that would explain the paradox goes
as follows: The United States has a special advantage in producing new products or
goods made with innovative technologies, such as aircraft and sophisticated computer
chips. Such products may well be less capital-intensive than products whose technology
has had time to mature and become suitable for mass production techniques. Thus
the United States may be exporting goods that heavily use skilled labor and innovative
entrepreneurship, while importing heavy manufactures (such as automobiles) that use
large amounts of capital.
Tests on Global Data Since the United States may be a special case, economists have
also attempted to broaden the test to incorporate more countries, as well as more factors of
production. An important such study by Harry P. Bowen, Edward E. Leamer, and Leo
Sveikauskas13 extended the predictions for the factor content of trade to 27 countries and
12 factors of production. The theory behind the test is the same as for Leontief’s test for
the United States: Based on the factor content of exports and imports, a country should be
a net exporter of a factor of production with which it is relatively abundantly endowed
(and conversely, net importer of those with which it is relatively poorly endowed).
12Later studies point to the disappearance of the Leontief paradox by the early 1970s. For example, see Robert
M. Stern and Keith E. Maskus, “Determinants of the Structure of U.S. Foreign Trade, 1958–76,” Journal of
International Economics 11 (May 1981), pp. 207–224. These studies show, however, the continuing importance
of human capital in explaining U.S. exports.
13See Harry P. Bowen, Edward E. Leamer, and Leo Sveikauskas, “Multicountry, Multifactor Tests of the Factor
Abundance Theory,” American Economic Review 77 (December 1987), pp. 791–809.
TABLE 5-2 Factor Content of U.S. Exports and Imports for 1962
Imports Exports
Capital per million dollars $2,132,000 $1,876,000
Labor (person-years) per million dollars 119 131
Capital-labor ratio (dollars per worker) $17,916 $14,321
Average years of education per worker 9.9 10.1
Proportion of engineers and scientists in work force 0.0189 0.0255
Source: Robert Baldwin, “Determinants of the Commodity Structure of U.S. Trade,” American Economic
Review 61 (March 1971), pp. 126–145.
100 PART ONE International Trade Theory
Table 5-3 shows one of the key tests of Bowen et al. The authors calculated the ratio of
each country’s endowment of each factor to the world supply of that factor. They then
compared these ratios with each country’s share of world income. If the factor-proportions
theory was right, a country would always export factors for which the factor share
exceeded the income share, and import factors for which it was less. In fact, for two-thirds
of the factors of production, trade ran in the predicted direction less than 70 percent of the
time. This result confirms the Leontief paradox on a broader level: Trade often does not
run in the direction that the Heckscher-Ohlin theory predicts. As with the Leontief paradox
for the United States, explanations for this result have centered on the failure of the
common technology assumption.
The Case of the Missing Trade Another indication of large technology differences
across countries comes from discrepancies between the observed volumes of trade and
those predicted by the Heckscher-Ohlin model. In an influential paper, Daniel Trefler14 at
the University of Toronto pointed out that the Heckscher-Ohlin model can also be used to
derive predictions for a country’s volume of trade based on differences in that country’s
factor abundance with that of the rest of the world (since, in this model, trade in goods is
substituting for trade in factors). In fact, factor trade turns out to be substantially smaller
than the Heckscher-Ohlin model predicts.
A large part of the reason for this disparity comes from a false prediction of largescale
trade in labor between rich and poor nations. Consider the United States, on one
side, and China on the other. In 2008, the United States had about 23 percent of world
income but only about 5 percent of the world’s workers; so a simple factor-proportions
theory would suggest that U.S. imports of labor embodied in trade should have been
huge, something like four times as large as the nation’s own labor force. In fact,
calculations of the factor content of U.S. trade showed only small net imports of labor.
Conversely, China had 7 percent of world income but approximately 20 percent of
TABLE 5-3 Testing the Heckscher-Ohlin Model
Factor of Production Predictive Success*
Capital 0.52
Labor 0.67
Professional workers 0.78
Managerial workers 0.22
Clerical workers 0.59
Sales workers 0.67
Service workers 0.67
Agricultural workers 0.63
Production workers 0.70
Arable land 0.70
Pasture land 0.52
Forest 0.70
*Fraction of countries for which net exports of factor runs in predicted direction.
Source: Harry P. Bowen, Edward E. Leamer, and Leo Sveikauskas, “Multicountry, Multifactor Tests of
the Factor Abundance Theory,” American Economic Review 77 (December 1987), pp. 791–809.
14Daniel Trefler, “The Case of the Missing Trade and Other Mysteries,” American Economic Review 85
(December 1995), pp. 1029–1046.
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 101
the world’s workers in 2008; it therefore “should” have exported most of its labor via
trade—but it did not.
Allowing for technology differences also helps to resolve this puzzle of “missing
trade.” The way this resolution works is roughly as follows: If workers in the United States
are much more efficient than those in China, then the “effective” labor supply in the
United States is much larger compared with that of China than the raw data suggest—and
hence the expected volume of trade between labor-abundant China and labor-scarce
America is correspondingly less.
If one makes the working assumption that technological differences between countries
take a simple multiplicative form—that is, that a given set of inputs produces only times
as much in China as it does in the United States, where is some number less than 1—it is
possible to use data on factor trade to estimate the relative efficiency of production in different
countries. Table 5-4 shows Trefler’s estimates for a sample of countries; they suggest
that technological differences are in fact very large. However, this exercise does not
prove that technology differences do have this simple multiplicative form. If they don’t,
then some country could have bigger technological advantages in particular sectors, and
the predictions for the pattern of trade would be a mix between those of the Ricardian and
Hecksher-Ohlin models.
Patterns of Exports Between Developed
and Developing Countries
Although the overall pattern of international trade does not seem to be very well accounted
for by a pure Heckscher-Ohlin model, comparisons of the exports of labor-abundant, skillscarce
nations in the third world with the exports of skill-abundant, labor-scarce nations do
fit the theory quite well. Consider, for example, Figure 5-12, which compares the pattern
of U.S. imports from Bangladesh, whose work force has low levels of education, with the
pattern of U.S. imports from Germany, which has a highly educated labor force.
In Figure 5-12, which comes from the work of John Romalis of the University of
Chicago,15 goods are ranked by skill intensity: the ratio of skilled to unskilled labor used
in their production. The vertical axes of the figure show U.S. imports of each good from
Germany and Bangladesh, respectively, as a share of total U.S. imports of that good. As
you can see, Bangladesh tends to account for a relatively large share of U.S. imports of
low-skill-intensity goods such as clothing, but a low share of highly skill-intensive goods.
Germany is in the reverse position.
d
d
TABLE 5-4 Estimated Technological Efficiency, 1983 (United States = 1)
Country
Bangladesh 0.03
Thailand 0.17
Hong Kong 0.40
Japan 0.70
West Germany 0.78
Source: Daniel Trefler, “The Care of the Missing Trade and Other Mysteries,” American Economic Review
85 (December 1995), pp. 1029–1046.
15John Romalis, “Factor Proportions and the Structure of Commodity Trade,” American Economic Review 94
(March 2004), pp. 67–97.
102 PART ONE International Trade Theory
Changes over time also follow the predictions of the Heckscher-Ohlin model. Figure 5-13
shows the changing pattern of exports to the United States from Western Europe, Japan, and
the four Asian “miracle” economies—South Korea, Taiwan, Hong Kong, and Singapore—
which moved rapidly from being quite poor economies in 1960 to relatively rich
economies with highly skilled work forces today.
Panel (a) of Figure 5-13 shows the pattern of exports from the three groups in 1960; the
miracle economies were clearly specialized in exports of low-skill-intensity goods, and
even Japan’s exports were somewhat tilted toward the low-skill end. As shown in panel
(b), by 1998, however, the level of education of Japan’s work force was comparable to that
of Western Europe, and Japan’s exports reflected that change, becoming as skill-intensive
as those of European economies. Meanwhile, the four miracle economies, which had rapidly
increased the skill levels of their own work forces, had moved to a trade pattern comparable
to that of Japan a few decades earlier.
A key prediction of the Heckscher-Ohlin model is that changes in factor abundance
lead to biased growth toward sectors that use that factor intensively in production. We can
see that the experience of those Asian economies fit very well with these predictions: As
the supply of skilled labor increased, they increasingly specialized in the production of
skill-intensive goods.
Implications of the Tests
We have just seen that the empirical testing of the Heckscher-Ohlin model has produced
mixed results. In particular, the evidence is weak concerning the prediction of the model
that, absent technology differences between countries, trade in goods is a substitute for
trade in factors: The factor content of a country’s exports does not always reflect that
0.06
0.08
0.10
0.12
Estimated share of US imports
by industry
Estimated share of US imports
by industry
0.004
0.003
0.002
0.001
0.000
0.04
0.02
Germany
(left scale)
Bangladesh
(right scale)
Skill intensity of industry
0.00
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Figure 5-12
Skill Intensity and the Pattern of U.S. Imports from Two Countries
Source: John Romalis, “Factor Proportions and the Structure of Commodity Trade,” American Economic Review 94
(March 2004), pp. 67–97.
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 103
country’s abundant factors; and the volume of trade is substantially lower than what would
be predicted based on the large differences in factor abundance between countries.
However, the pattern of goods trade between developed and developing countries fits the
predictions of the model quite well.
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Share of U.S. imports by industry
Four miracles
Japan
Western Europe
Skill intensity of industry
0.05
(a) 1960
(a) 1998
0.10 0.15
1960
1998
0.20 0.25 0.30 0.35 0.40
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Share of U.S. imports by industry
Four miracles
Japan
Western Europe
Skill intensity of industry
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Figure 5-13
Changing Patterns of Comparative Advantage
104 PART ONE International Trade Theory
The Heckscher-Ohlin model also remains vital for understanding the effects of trade,
especially its effects on the distribution of income. Indeed, the growth of North-South
trade in manufactures—a trade in which the factor intensity of the North’s imports is very
different from that of its exports—has brought the factor-proportions approach into the
center of practical debates over international trade policy.
SUMMARY
1. To understand the role of resources in trade, we develop a model in which two goods
are produced using two factors of production. The two goods differ in their factor
intensity, that is, at any given wage-rental ratio, production of one of the goods will use
a higher ratio of capital to labor than production of the other.
2. As long as a country produces both goods, there is a one-to-one relationship between the
relative prices of goods and the relative prices of factors used to produce the goods. A rise
in the relative price of the labor-intensive good will shift the distribution of income in
favor of labor, and will do so very strongly: The real wage of labor will rise in terms of
both goods, while the real income of capital owners will fall in terms of both goods.
3. An increase in the supply of one factor of production expands production possibilities,
but in a strongly biased way: At unchanged relative goods prices, the output of the
good intensive in that factor rises while the output of the other good actually falls.
4. A country that has a large supply of one resource relative to its supply of other
resources is abundant in that resource. A country will tend to produce relatively more
of goods that use its abundant resources intensively. The result is the basic Heckscher-
Ohlin theory of trade: Countries tend to export goods that are intensive in the factors
with which they are abundantly supplied.
5. Because changes in relative prices of goods have very strong effects on the relative
earnings of resources, and because trade changes relative prices, international trade
has strong income distribution effects. The owners of a country’s abundant factors gain
from trade, but the owners of scarce factors lose. In theory, however, there are still
gains from trade, in the limited sense that the winners could compensate the losers,
and everyone would be better off.
6. In an idealized model, international trade would actually lead to equalization of the
prices of factors such as labor and capital between countries. In reality, complete
factor-price equalization is not observed because of wide differences in resources, barriers
to trade, and international differences in technology.
7. Empirical evidence is mixed on the Heckscher-Ohlin model, but most researchers
do not believe that differences in resources alone can explain the pattern of world
trade or world factor prices. Instead, it seems to be necessary to allow for substantial
international differences in technology. Nonetheless, the Heckscher-Ohlin
model does a good job of predicting the pattern of trade between developed and
developing countries.
KEY TERMS
abundant factor, p. 91
biased expansion of production
possibilities, p. 88
equalization of factor
prices, p. 97
factor abundance, p. 80
factor intensity, p. 80
factor prices, p. 85
factor-proportions
theory, p. 80
Heckscher-Ohlin theory, p. 80
Leontief paradox, p. 98
scarce factor, p. 91
skill-biased technological
change, p. 95
CHAPTER 5 Resources and Trade: The Heckscher-Ohlin Model 105
PROBLEMS
1. Go back to the numerical example with no factor substitution that leads to the production
possibility frontier in Figure 5-1.
a. What is the range for the relative price of cloth such that the economy produces
both cloth and food? Which good is produced if the relative price is outside of this
range?
For parts (b) through (f), assume that the price range is such that both goods are
produced.
b. Write down the unit cost of producing one yard of cloth and one calorie of food as
a function of the price of one machine-hour, r, and one work-hour, w. In a competitive
market, those costs will be equal to the prices of cloth and food. Solve for the
factor prices r and w.
c. What happens to those factor prices when the price of cloth rises? Who gains and
who loses from this change in the price of cloth? Why? Do those changes conform
to the changes described for the case with factor substitution?
d. Now assume that the economy’s supply of machine-hours increases from 3,000 to
4,000. Derive the new production possibility frontier.
e. How much cloth and food will the economy produce after this increase in its
capital supply?
f. Describe how the allocation of machine-hours and work-hours between the cloth
and food sectors changes. Do those changes conform with the changes described
for the case with factor substitution?
2. In the United States, where land is cheap, the ratio of land to labor used in cattle
raising is higher than that of land used in wheat growing. But in more crowded
countries, where land is expensive and labor is cheap, it is common to raise cows
by using less land and more labor than Americans use to grow wheat. Can we still
say that raising cattle is land-intensive compared with farming wheat? Why or
why not?
3. “The world’s poorest countries cannot find anything to export. There is no resource
that is abundant—certainly not capital or land, and in small poor nations not even
labor is abundant.” Discuss.
4. The U.S. labor movement—which mostly represents blue-collar workers rather than
professionals and highly educated workers—has traditionally favored limits on
imports from less-affluent countries. Is this a shortsighted policy or a rational one in
view of the interests of union members? How does the answer depend on the model
of trade?
5. Recently, computer programmers in developing countries such as India have begun
doing work formerly done in the United States. This shift has undoubtedly led to
substantial pay cuts for some programmers in the United States. Answer the following
two questions: How is this possible, when the wages of skilled labor are
rising in the United States as a whole? What argument would trade economists
make against seeing these wage cuts as a reason to block outsourcing of computer
programming?
6. Explain why the Leontief paradox and the more recent Bowen, Leamer, and Sveikauskas
results reported in the text contradict the factor-proportions theory.
7. In the discussion of empirical results on the Heckscher-Ohlin model, we noted
that recent work suggests that the efficiency of factors of production seems to differ
internationally. Explain how this would affect the concept of factor-price
equalization.
106 PART ONE International Trade Theory
FURTHER READINGS
Donald R. Davis and David E. Weinstein. “An Account of Global Factor Trade.” American
Economic Review 91 (December 2001), pp. 1423–1453. The authors review the history of tests of
the Heckscher-Ohlin model and propose a modified version—backed by extensive statistical
analysis—that allows for technology differences, specialization, and transportation costs.
Alan Deardorff. “Testing Trade Theories and Predicting Trade Flows,” in Ronald W. Jones and Peter
B. Kenen, eds. Handbook of International Economics. Vol. 1. Amsterdam: North-Holland, 1984.
A survey of empirical evidence on trade theories, especially the factor-proportions theory.
Gordon Hanson and Ann Harrison. “Trade and Wage Inequality in Mexico.” Industrial and Labor
Relations Review 52 (1999), pp. 271–288. A careful study of the effects of trade on income
inequality in our nearest neighbor, showing that factor prices have moved in the opposite direction
from what one might have expected from a simple factor-proportions model. The authors
also put forward hypotheses about why this may have happened.
Ronald W. Jones. “Factor Proportions and the Heckscher-Ohlin Theorem.” Review of Economic
Studies 24 (1956), pp. 1–10. Extends Samuelson’s 1948–1949 analysis (cited below), which
focuses primarily on the relationship between trade and income distribution, into an overall
model of international trade.
Ronald W. Jones. “The Structure of Simple General Equilibrium Models.” Journal of Political
Economy 73 (December 1965), pp. 557–572. A restatement of the Heckscher-Ohlin-Samuelson
model in terms of elegant algebra.
Ronald W. Jones and J. Peter Neary. “The Positive Theory of International Trade,” in Ronald W.
Jones and Peter B. Kenen, eds. Handbook of International Economics. Vol. 1. Amsterdam:
North-Holland, 1984. An up-to-date survey of many trade theories, including the factorproportions
theory.
Bertil Ohlin. Interregional and International Trade. Cambridge: Harvard University Press, 1933.
The original Ohlin book presenting the factor-proportions view of trade remains interesting—its
complex and rich view of trade contrasts with the more rigorous and simplified mathematical
models that followed.
Robert Reich. The Work of Nations. New York: Basic Books, 1991. An influential tract that argues
that the increasing integration of the United States in the world economy is widening the gap
between skilled and unskilled workers.
John Romalis. “Factor Proportions and the Structure of Commodity Trade.” The American
Economic Review 94 (March 2004), pp. 67–97. A recent, state-of-the-art demonstration that a
modified version of the Heckscher-Ohlin model has a lot of explanatory power.
Paul Samuelson. “International Trade and the Equalisation of Factor Prices.” Economic Journal 58
(1948), pp. 163–184; and “International Factor Price Equalisation Once Again.” Economic
Journal 59 (1949), pp. 181–196. The most influential formalizer of Ohlin’s ideas is Paul
Samuelson (again!), whose two Economic Journal papers on the subject are classics.
No comments:
Post a Comment