There is a common misconception that China’s growth is taking place at the expense of its many trading partners. This has prompted threats of trade policy retaliation on the part of many of the trading partners, not the least of which is the United States. A useful overview of the Chinese role in regional growth and development was provided by Phillip Day.1 He correctly points out that even though exports of other Asian countries to the United States fell as those of China increased, the total exports of those other countries grew in a complementary fashion through increased trade with each other. The reason is that China was already the largest importer of South Korean and Taiwanese goods as well as a substantial importer from Japan (if exports into Hong Kong are taken into account). Interestingly, as China grew, it found itself on the midpoint of a supply chain in that it imported high-tech components from East Asia, assembled them into final commodities, and exported them to final end-markets throughout the world. Thus, instead of hurting other countries in the region, China’s rapid growth and emergence as an export powerhouse in the world economy had a positive impact on other East Asian countries. Unfortunately, the politicians, trade groups, and companies that were critical of China’s export success also ignored the fact that it was often foreign investment and foreign companies that underpinned the Chinese export success. China’s notable rate of growth in recent years and its growing impact on world trade and globalization reflect the fact that the production possibilities for a country do not remain fixed and are often fostered by the country’s economic policies. Growth in output potential is represented by outward shifts in the production-possibilities frontier (PPF), which enable the country to reach a higher level of real income (a consumption-possibilities frontier further to the right) and presumably a higher level of well-being. Growth comes about by means of change in technology or through the acquisition of additional resources such as labor, physical capital, or human capital. Inasmuch as international trade affects and is affected by economic growth, it is important to examine several of the more important economic implications of growth. This chapter begins by pointing out how growth influences trade through changes in both production and consumption. This is followed by a discussion of the sources of growth and the manner in which they influence changes in the economy. The chapter concludes by looking briefly at the effect of growth on the country’s economic well-being when the country is participating in international trade.

Classifying the Trade Effects of Economic Growth

As real income increases, it affects both producers and consumers. Producers need to decide how to alter production, given the increase in resources or the change in technology. Consumers, on the other hand, are faced with how to spend the additional real income. Both of these decisions have implications for the country’s participation in international trade and thus for determining whether countries become more or less open to trade as economic growth occurs. We begin this analysis by categorizing the alternative production and consumption responses that accompany economic growth in terms of their respective implications for international trade.

Trade Effects of Production Growth

Let us assume that a small country is characterized by increasing opportunity costs and is currently in equilibrium at a given set of international prices (see Figure 1), remembering that a small country cannot influence world prices.2 In panel (a), France is producing at point A and consuming at point B. To do this, France exports wine and imports electronics. As growth occurs, the PPF shifts outward, and French producers have the opportunity to select a point on the new PPF that will maximize their profits. In general terms, they have the possibility of producing (1) more of both commodities in the same proportion as at point A, (2) more of both commodities but relatively more of one than the other, or (3) absolutely more of one commodity and absolutely less of the other. These possibilities can be demonstrated on our figure and will form the basis for categorizing the various production trade effects that can accompany growth.

To establish the classification of the trade effects of growth, return to point A. This will become the origin for new mini-axes, shown as dashed lines in panel (b). Points lying to the left of the dashed vertical line reflect cases where the new production of wine is less than at point A. Points to the right of this vertical line indicate cases where the new production of wine is greater than at point A. Similarly, points lying above the dashed horizontal mini-axis reflect greater production of electronics, whereas points below this line indicate less production of electronics. Points lying above and to the right of point A thus represent larger production of both goods. Production points lying on the straight line passing through the origin and point A reflect outputs of electronics and wine that are proportionally the same as at A; that is, the ratio of electronics to wine production is a constant. Points beyond point A that fall on this line demonstrate a neutral production effect because production of the export good and the import-competing good have grown at the same rate.page 206 The remaining production possibilities with growth conveniently fall into four regions, which are isolated by the neutral ray from the origin and the mini-axes at point A. Region I represents possible new production points that reflect increased production of both commodities, but where the change in the production of wine is relatively greater than the change in the production of electronics. Because wine is the export good, this type of growth has a protrade production effect, reflecting the relatively greater availability of the export good. Region II contains production-possibilities points that demonstrate increased production of wine but a decrease in production of electronics. New production points lying in this region as a result of growth fall in the ultra-protrade production effect category, suggesting an even greater potential effect on the desire to trade. New production points lying in region III reflect higher production levels of both goods but relatively greater increases in electronics than in wine. Because electronics are the import-competing good, growth reflecting this production change has an antitrade production effect. Finally, new production points lying in region IV, with increased production of electronics and less of wine, are placed in the ultra-antitrade production effect category. The actual point of production after growth will be the point where the new enlarged PPF is tangent to the international price line. This point will necessarily fall in one of the aforementioned regions.

Trade Effects of Consumption Growth

A similar technique can be used to describe the various consumption effects of growth. In this case, we analyze the nature of consumer response to growth relative to the original equilibrium at point B [Figure 1, panel (a)]. Figure 2 focuses on this initial equilibrium point, which serves as the origin for new mini-axes.

With growth there is an increase in real income indicated by the rightward shift in the consumption-possibilities line (the international terms-of-trade line). This allows consumers to choose combinations of electronics and wine previously not possible. The consumption effects of growth on trade can be isolated by the mini-axes whose origin is at pregrowth consumption point B. If the new consumption point is on the straight line from the origin through B, consumption of both goods will increase proportionally and the consumption trade effect will be neutral. Should the new consumption point fall in region I, it is an antitrade consumption effect; if it falls in region II, it is an ultra-antitrade consumption effect; if it falls in region III, it is a protrade consumption effect; and if it falls in region IV, it is an ultra-protrade consumption effect.page 207 Points lying to the left of the dashed vertical axis reflect less consumption of wine, while points to the right indicate greater consumption. Points lying below the dashed horizontal axis reflect less electronics consumption, while points above that line indicate more. Points lying beyond B on the straight line passing through point B and the origin of the original axes indicate cases where goods are consumed in the same proportion as at point B. Points so situated reflect a neutral consumption effect, because consumers have not changed their relative consumption pattern with growth. The remaining effects will be isolated in a manner similar to that used in the production analysis. New consumption points lying in region I as a result of growth in real income reflect a relatively larger increase in the consumption of wine than in that of electronics. Because wine is the export good, the change in consumption reduces the country’s relative willingness to export. This effect is called an antitrade consumption effect. An even more extreme case of this type of behavior is found in region II, where the consumption of wine increases and that of electronics falls. This response is called an ultra-antitrade consumption effect. If growth causes consumption to move into region III, where consumption of both goods increases but consumption of electronics (the import good) increases relatively more than wine, a protrade consumption effect occurs. Finally, if consumption of electronics increases and consumption of wine actually falls with growth (region IV), an ultra-protrade consumption effect exists.4 The ultimate impact of economic growth on trade depends on the effects on both production and consumption. The expansionary impact of growth on trade is larger whenever both the production and the consumption effects are in the “pro” or “ultra-pro” regions. The total effect of growth on trade is demonstrated with three different cases in Figure 3. In panel (a), both production and consumption effects are in the ultra-antitrade category. With growth, production moves from point A to point A′ and consumption from point B to point B′. Note that commodity prices are fixed, because this is a small country. The result of growth is a reduction in trade, reflected in the new trade triangle, A′R′B′, which is smaller than the original triangle, ARB.

In panel (b), the production effect is a protrade effect and the consumption effect is neutral. These effects can be observed in the position of point A′ and point B′ with respect to point A and point B. The result is a relative expansion of trade (trade triangle A′R′B′). In panel (c), an ultra-protrade production effect is coupled with a protrade consumption effect. Again, trade expands relatively (from ARB to A′R′B′). As we move from panel (a) to panel (c) in Figure 3, the new trading triangle gets successively larger. While the volume of trade generally increases with growth, this is not always true. For example, growth leading to ultra-antitrade consumption and production effects actually causes trade to decline. A useful way to summarize the net result of production and consumption effects on the growing country’s trade is through the concept of the income elasticity of demand for imports (YEM). This measure is the percentage change in imports divided by the percentage change in national income. If YEM = 1.0, then trade is growing at the same rate as national income, and the net effect is neutral. If 0 < YEM < 1, trade is growing in absolute terms but at a slower rate than income; the net effect is antitrade. If YEM < 0, trade is actually falling as income grows (ultra-antitrade effect). Finally, if YEM > 1.0 (imports or trade growing more rapidly than national income), there is a protrade or ultra-protrade net effect. (The algebraic distinction between protrade and ultra-protrade is more complex and need not concern us.) As a general rule, if both the production effect and the consumption effect are of the same type (e.g., both “protrade”), then the net or overall effect will be of the same type as the two individual effects. If one effect is protrade (antitrade) and the other is neutral, the net effect will be protrade (antitrade). There are obviously various other combinations, and some of them require more information on the precise size of each of the two effects before the net result can be determined, such as with a protrade production effect that is coupled with an antitrade consumption effect.

Sources of Growth and the Production-Possibilities Frontier

In the introduction to this chapter, we mentioned that growth can result from changes in technology or the accumulation of factors such as capital and labor. Because they affect the PPF in different ways, we will examine the two kinds of changes individually.

The Effects of Technological Change

Technological change alters the manner in which inputs are used to generate output, and it results in a larger amount of output being generated from a fixed amount of inputs. Let us assume that we are dealing with two inputs, capital and labor. The new technology may be factor neutral; that is, it results in the same relative amounts of capital and labor being used as before the technology changed (at constant factor prices). However, smaller amounts of inputs are used per unit of output. On the other hand, the new technology might be labor saving in nature. In this instance, fewer factor inputs are required per unit of output, but the relative amount of capital used rises at constant factor prices (i.e., the K/L page 209ratio increases). Finally, the technological change could lead to a decrease in the K/L ratio at constant factor prices. In this instance, we say that the technological change is capital saving. In effect, a labor- (capital-) saving technological change has an effect equivalent to increasing the relative amount of labor (capital) available to the economy. It is easy to see why a technological change that reduces the relative labor requirement per unit of output (labor-saving technological change) is not necessarily thought desirable in a relatively labor-abundant developing country. We limit our analysis of technological change to the factor-neutral type. On the PPF, a factor-neutral change in technology that affects one commodity means that the country is able to produce more of that commodity for all possible levels of output of the second commodity. Thus, this commodity-specific change in technology causes the PPF to move outward except at the intercept for the nontechnology-changing commodity (see Figure 5). In panel (a) of Figure 5, if the change in technology occurs in autos, this is shown by the PPF that is the farthest out along the autos axis. On the other hand, the PPF that is the farthest out along the food axis indicates what happens if the technological change occurs only in food production. Finally, if the change in technology affects both commodities in the same relative manner, the PPF shifts outward in an equiproportional fashion, as demonstrated in panel (b) of Figure 5. This is commodity-neutral technological change.

Labour and Capital Requirements per unit of Output

Figure 4 indicates the changes in the relative use of capital and labor that took place in six countries from the mid-1960s to the mid-1980s. The changes are measured per unit of output. The three points on each graph show the actual level of capital and labor used, and the isoquants demonstrate the nature of substitution between capital and labor in the country for each year. Although the nature of the adjustment has been different in the six countries, the use of capital relative to labor has clearly increased in all of them. Japan and Germany experienced the greatest increase in the K/L ratio, and the U.S. ratio appears to have increased the least.

Traditionally, technological change has been treated exogenously (i.e., as an independent event from outside) in the growth literature, often at a fixed rate of growth.5 However, in the late 1980s a series of long-run growth models began to appear in which the rate of technological change was determined endogenously, or within the system, instead of being imposed from outside. In these newer models, the rate of technological change is determined by such factors as the growth in physical capital and the increase in human capital. New investment fosters and/or embodies new innovations and inventions which can, in turn, stimulate additional technological change as experience with the new capital leads to more change in a “learning-by-doing” environment. Similar “spillover” effects are also linked to the acquisition of human capital as well as to expenditures on research and development (R&D). These models, generally referred to as endogenous growth models, reflect the basic idea that change in technology is the result of things that people do, not something produced outside a particular economic system.6 In so doing, they have provided an explanation of how rapid sustainable growth can take place, avoiding the traditional neoclassical conclusion that economic growth would ultimately converge to the natural rate of population growth due to the declining productivity of capital. Grossman and Helpman (1991) importantly added to the literature on endogenous growth by examining the implications of endogenous technological change for international issues including dynamic comparative advantage, trade and growth, product cycles, and the international transmission of policies. More recent literature has focused specifically on research and development as the key factor in endogenous growth models. A distinction is further made between “first-generation” and “second-generation” endogenous growth models. In the former, the growth rate of total factor productivity is proportional to the number of R&D workers; in the latter, modifications are made to this assumption (such as diminishing returns to the number of R&D workers). See Madsen, Saxena, and Ang (2010). We do not pursue these developments in this book, however. For our purposes, whether technological change is exogenous or endogenous, it still results in an outward shift of the PPF. page 212 IN THE REAL WORLD: “SPILLOVERS” AS A CONTRIBUTOR TO ECONOMIC GROWTH When a country’s trading partners experience economic growth, it is clear that such growth can have effects on the growth of any given domestic country, as noted in the opening vignette to this chapter regarding China. The effects can occur, for example, because the partners increase their imports with growth—which in turn stimulates income in the domestic country because of the boost in its exports—and because the growing trading partners may transfer capital and technology abroad through engaging in more foreign direct investment. Two International Monetary Fund economists, Vivek Arora and Athanasios Vamvakidis, have attempted to provide quantitative estimates of such country “spillover” effects.* Using data for 101 developed and developing countries over the 1960–1999 period, they sought to statistically explain countries’ economic growth rates using traditional variables such as investment in physical capital, investment in human capital, and general openness to international trade. However, Arora and Vamvakidis also included the real per capita gross domestic product (GDP) growth of trading partners as a variable in their regression analysis, as well as the ratio of a domestic country’s real per capita GDP to the country’s trading partners’ real per capita GDP. These last two variables were designed to isolate the spillover effects of trading partners’ growth on a domestic country’s growth. The fact that such spillovers are important in the real world was clearly confirmed. First, Arora and Vamvakidis estimated that after controlling for other determinants of a country’s growth, a 1 percentage point increase in the growth rate of a domestic country’s trading partners was associated with a 0.8 percentage point increase in the growth rate of the domestic country. Further, this positive impact had increased over time because the spillover was larger during the 1980–1999 period than it was for the 1960–1999 period as a whole. Faster growth in trading partners can obviously lead to a greater growth in demand for the domestic country’s exports, for example. A second important result was that a developing country’s growth rate, after controlling for other factors, was negatively correlated with the closeness of the level of that country’s real per capita GDP to the average real per capita GDP of its trading partners. Stated another way, a developing country with a very low per capita income that is trading mostly with high-income countries will receive a greater spillover effect than it would if its income were more similar to the incomes of its trading partners. This greater spillover effect could reflect greater opportunities to make larger leaps in technological advance through transfer of the high technology of the trading partners through foreign direct investment. An implication of this finding is that as a country’s income level approaches that of its trading partners, the country’s growth rate, other things equal, will slow down. *Vivek Arora and Athanasios Vamvakidis, “Economic Spillovers,” Finance and Development 42, no. 3 (September 2005), pp. 48–50.

The Effects of Factor Growth

The second source of economic growth is increased availability of the factors of production. We consider the impact of factor growth in terms of two homogeneous inputs, capital and labor. In the real world, there are other primary inputs such as natural resources, land, and human capital, and factors do not tend to be homogeneous. Labor and capital remain, however, two of the most important inputs, and the insights gained from examining K and L can be extended to the more general case. Estimates for the growth in capital and labor for selected countries for 1966–2014 are presented in Table 1.

An increase in factor abundance can take place through increases in capital stock, increases in the labor force, or both. The capital stock of a country grows as domestic and foreign investment occurs in the country. The labor force expands through increases in population (including immigration), increases in the labor force participation rate, or both. If both labor and capital grow at the same rate, the PPF will shift out equiproportionally, as in the case of commodity-neutral technological change. This factor-neutral growth effect is demonstrated in panel (a) in Figure 6 with the new PPF that is farther out than the old.

The matter is more complex if one of the factors grows and the other does not. Suppose that the capital stock increases but the size of the labor force remains constant. How will the PPF change? In answering this question, remember the production assumptions from neoclassical theory and the Heckscher-Ohlin analysis. Assume that cutlery is capital intensive and cheese is labor intensive. If the capital stock grows, it has the greatest relative impact on the capital-intensive product. Think of this as an expansion of the Edgeworth box (see Chapter 5) along the capital side, with the labor side remaining the same size. If all the country’s resources are devoted to the production of cutlery, the expansion of the capital stock permits the country to reach a higher output level (higher isoquant) than that reached prior to the growth of capital. The growth in capital also permits a larger amount of cheese to be produced for any level of cutlery because capital can be substituted to some degree for labor. However, because cheese is the labor-intensive good, the potential impact on production is less than it is for the capital-intensive good. Consequently, the PPF shifts outward asymmetrically in the direction of the capital-intensive good. This shift is demonstrated in panel (b) of Figure 6. An analogous argument can be made for growth in labor when the capital stock is held constant. Then the PPF shifts outward in an asymmetrical manner, with the labor-intensive product showing a greater relative response. The effect of growth in the labor force is demonstrated in Figure 6(c).

Factor Growth, Trade, and Welfare in the Small-Country Case

A non-neutral growth in factors will shift the PPF in an asymmetrical manner and alter the relative factor abundance in the country. The economic response to this change depends on relative commodity prices. Let us continue to assume that the country is a small country and cannot influence world prices, which remain constant. What happens to production in this case when one factor, labor, for example, grows and capital stock remains fixed? We already know that the PPF will shift outward relatively more along the axis of the labor-intensive commodity. When this occurs, production takes place at the point of tangency between the new PPF and the same set of relative prices (see Figure 7). This new tangency occurs at a level of production that represents an increase in output of the labor-intensive good and a decrease in output of the capital-intensive good. If the labor-intensive good is the export good, this is an ultra-protrade production effect; if the labor-intensive good is the import good, growth in labor produces an ultra-antitrade production effect. The conclusion that growth in one factor leads to an absolute expansion in the product that uses that factor intensively and an absolute contraction in output of the product that uses the other factor intensively is referred to as the Rybczynski theorem after the British economist T. M. Rybczynski. The economics that lie behind the Rybczynski theorem is straightforward. Because, by the small-country assumption, relative product prices cannot change, then relative factor prices cannot change because technology is constant. If relative factor prices are unchanged in the new equilibrium, then the K/L ratios in the two industries at the new equilibrium are the same as before the growth. The only way this can happen, given the increased amount of labor, is if the capital-intensive sector releases some of its capital to be used with the new labor in the labor-intensive sector. When this transfer of capital occurs, output of the capital-intensive good falls and output of the labor-intensive good expands.

What effect does factor growth have on trade in the small-country case? The production impact of factor growth on trade depends on whether the growing factor (labor in our example) is the abundant or the scarce factor. If it is the abundant factor, there is an ultra-protrade production effect, assuming the country is exporting the commodity that is intensive in the abundant factor, in the manner of Heckscher-Ohlin. If it is the scarce factor, there is an ultra-antitrade production effect. Other things being equal, therefore, the expansionary impact on trade is greater with growth in the abundant factor than in the scarce factor. The total effect on trade depends on both production and consumption effects, however. As a general rule, if the consumption effect is protrade, then the country will participate more heavily in trade if the abundant factor grows. If the scarce factor grows, the total effect can be less participation in trade. A full assessment of the impacts of factor growth on the country’s participation in trade requires estimation of both supply and demand effects. Consider the effect of growth on welfare. If capital grows or there is technological change, there is an increase in well-being, because either of those changes will increase real per capita income and permit the country to reach a higher community indifference curve. It is assumed that the social benefits resulting from the increased output are not accompanied by an increase in income inequality. However, if there is growth in the labor force, the welfare implications of growth are less straightforward. The community indifference curve map that existed prior to growth is no longer relevant, because the new members of the labor force may have different tastes than the original members. It is, therefore, not possible to use the two different indifference curve maps to make welfare comparisons. In practice, economists use levels of per capita income to approximate changes in country welfare. While this measure has deficiencies, it appears to correlate well with many other variables indicative of welfare. It does not, however, take explicit account of changes in income distribution. If we adopt per capita income as the measure of welfare in the case of labor force growth, what can be concluded about the impact of such growth on welfare? We have assumed that our production is characterized by using two inputs and that there are constant returns to scale. The definition of constant returns to scale states that if all inputs increase by a given percentage, output will increase by the same percentage. If, however, only one input expands, output will expand by a smaller percentage than the increase in the single factor. (See Concept Box 1 for additional discussion of this point.) Thus, if we use per capita income as our measure of well-being, we conclude that an increase in population (labor) will lead to a fall in per capita income and hence in country well-being, other things being equal.

Growth, Trade, and Welfare: The Large Country Case

The effects of growth on trade to this point have been based on the assumption that the country cannot influence the international terms of trade. However, a country could influence world prices of a commodity if the country is a sufficiently large consumer or producer. In that instance, we must also take into account the possible effects of economic growth on the terms of trade. Suppose we are dealing with a large country that can influence international prices and that growth of the abundant factor, in this case capital, causes an ultra-protrade production effect. Assume further that this is coupled with a neutral consumption effect. The total effect on trade is that this country demands more imports and supplies more exports at the current set of international prices [see panel (a) of Figure 9]. As a result of growth, this country alters its “offer” at that particular set of prices on the world market. The increased supply of the export good (good B) and the increased demand for the import good (good A) reduce the international terms of trade [see panel (b) of Figure 9]. (For a discussion of how the different types of growth affect the offer curve of a growing country, see Concept Box 2.) The increase in the relative price of imports effectively reduces the possible gains from growth and trade, because the country now receives fewer imports per unit of exports (see Figure 11, page 220). Graphically, the international terms-of-trade line TOT1 is flatter now than before growth (TOT0), and it is tangent to a lower indifference curve (IC2) than would be the case if prices had not been affected (IC1). Thus, some of the gains of growth are effectively offset by the deterioration in the terms of trade. For growth to be beneficial to the large trading country, these negative terms-of-trade effects must not completely offset the positive effects of growth.

Concept Box 1: Labour Force Growth and Per Capita Income

Under the assumption of constant returns to scale, a 20 percent growth in the labor force leads to a 20 percent growth rate in the output of a particular commodity only if all other inputs also grow at 20 percent. If all inputs grow by a fixed percentage, the PPF MN shifts out to PPF M ′N ′ in an equidistant manner by a similar percentage, as shown in Figure 8. However, if only labor grows, the PPF shifts out relatively more for the labor-intensive good than for the capital-intensive good, as indicated by the dashed PPF M ″N ″. But, because only labor is growing, the outward shift from MN to M ″N ″ must be less for all combinations of the two final goods than was the case when all inputs and output increased by the same percentage. It follows that whatever the combination of the country’s two products, the increase in income represented by M ″N ″ is always less than that represented by M ′N ′, other things being equal. Thus, a 20 percent increase in the labor force leads to an increase in income that is less than 20 percent, and per capita income therefore declines.

Growth can result in declining well-being in two ways in the large-country case. First, if labor is the abundant and growing resource, the loss in welfare linked to the resulting decline in per capita income is further augmented by the deterioration in the international terms of trade (increase in the relative price of imports). The result is essentially the same as in the small-country case except it is intensified by the negative terms-of-trade effect. Second, even if capital is the growing abundant factor (or there is technological change in the export commodity) and the negative terms-of-trade effects are sufficiently strong, the country could be worse off after growth (see Figure 12, page 220). In this case, the deterioration in the terms of trade is so great that the new, flatter international terms-of-trade line (TOT1) is tangent to a lower community indifference curve (IC2 at point C2) than it was prior to growth (IC0 at point C0). When the negative terms-of-trade effects outweigh the positive growth effects in this manner, the situation is referred to as immiserizing growth, first pointed out by Jagdish N. Bhagwati (1958). We need to discuss briefly the effects of growth in the scarce factor for a large country. According to the Rybczynski theorem, growth in the scarce factor leads to an increase in output of the import-competing good and a decrease in output of the export good. Ignoring any offsetting consumption effects, for the large country this leads to a reduction in the “offer” of exports for imports by the expanding country since growth is ultra-antitrade biased [see panel (b) of Figure 13, page 221]. The growth phenomenon leads to an improvement in the terms of trade faced by this country, as the reduced amount of exports places upward pressure on the price of the export good and the reduced import demand produces downward pressure on the price of the import good. The positive effects of growth are enhanced by the terms-of-trade effects, causing the country to reach an even higher indifference curve. This effect is shown in panel (a) of Figure 13, as consumer welfare rises from IC0 before growth to IC1 with growth alone to IC2 after the terms-of-trade effects are taken into account. Finally, if labor is the growing scarce factor, the positive terms-of-trade effects can offset, at least in part, some of the loss in well-being due to declining per capita income.

Concept Box 2: Economic Growth and the Offer Curve

When a country experiences economic growth, its offer curve will shift. However, the extent and even the direction of the shift depend on the type of growth that takes place. (See Meier, 1968, p. 18.) In Figure 10, the pregrowth offer curve of country I is OCI. If the net effect of country I’s production and consumption effects is ultra-protrade growth, its offer curve shifts rightward to OCUP, with a consequent increase in the volume of trade and a deterioration of the terms of trade with trading partner country II. If the net effect is protrade growth, the offer curve of country I shifts to OCP, with a smaller increase in the volume of trade and a smaller deterioration in the terms of trade than with OCUP. But, perhaps surprisingly, even with neutral growth, country I’s offer curve still shifts to the right (to OCN). This result occurs because, even though trade in relation to national income for country I has remained constant (since the income elasticity of demand for imports, YEM, equals 1.0), the absolute willingness to trade increases. That the absolute amount of trade increases is also true even with antitrade-biased growth (offer curve OCA), despite the fact that trade is falling relative to national income (0 < YEM < 1). Finally, a net effect of ultra-antitrade-biased growth shifts the offer curve of country I leftward to OCUA. Only in this case, other things equal, will the volume of trade decrease and the terms of trade improve.

Growth and the Terms of Trade: A Developing-Country Perspective

The preceding analysis of growth, trade, and welfare provides a useful background for examining the interaction among growth, trade, and economic development. The importance of technological change and the accumulation of capital in improving country welfare is certainly clear. In countries where population and thus labor is growing at a relatively high rate, some stimulus to production in addition to labor must occur if per capita incomes are to improve steadily. It is also important to consider the possible effect of growth on the international terms of trade. Although most developing countries are not large in an overall economic sense, many are sufficiently important suppliers of individual primary commodities to be able to influence world prices. Several important observations need to be made. First, economic growth based on expansion of production of these goods may well lead to adverse terms-of-trade movements. Although immiserizing growth does not appear to be common in the real world, adverse terms-of-trade movements clearly reduce the benefits of growth and trade to the developing countries. This observation provides strong support for considering product diversification in the development strategy to reduce the likelihood of growth contributing to negative terms-of-trade movements and the reliance on only one main product for export earnings. A major world supplier of an export good such as coffee, cocoa, or groundnuts that relies heavily on the particular commodity for its export proceeds could find itself in difficult economic and financial straits if a bumper crop drives down world prices.

Terms of Trade of Brazil, Jordan, Pakistan, and Thailand, 1980-2014

The terms-of-trade behavior of four developing countries—Brazil, Jordan, Pakistan, and Thailand—during recent decades is presented in Figure 14. The graph indicates that the greatest deterioration in the terms of trade over this time period occurred for Pakistan and Brazil. In Pakistan, there was an overall rise in its terms of trade in the 1980s, general stability in the 1990s, and then a rather sharp decline for the rest of the period. Pakistan had a relatively large average annual GDP growth rate of 6.3 percent from 1980 to 1990 and average growth rates of 3.8 percent from 1990 to 2000, 5.1 percent from 2000 to 2009, 3.1 percent from 2009 to 2013, and 5.4 percent in 2014. The experience in the 1990s and in the first years of the 2000s shows some consistency with the expectations from this chapter in that sluggish GDP growth in the 1990s yielded little change in the terms of trade while faster growth in the 2000s yielded terms-of-trade deterioration. However, the rapid GDP growth of the 1980s would not, on the basis of the analysis in this chapter, be expected to be associated with an improvement in the terms of trade. For Brazil, there was an improvement in its terms of trade in the 1980s and until 1992, followed by a sharp drop through 2001, an upward spike for two years, and a decline through 2014. Brazil’s annual average GDP growth rate was 2.7 percent from 1980 to 1990 and again from 1990 to 2000, 3.6 percent from 2000 to 2009, 3.1 percent from 2009 to 2013, and a miniscule 0.1 percent in 2014. Because the average GDP growth rates in general varied little for Brazil while Brazil’s terms of trade changed dramatically, other factors besides growth of GDP were clearly at work on the country’s terms of trade. Thailand also experienced a rather large terms-of-trade decline over the 1980–2014 period as a whole. The decline also appears to have occurred regularly throughout that extended period. Thailand’s annual average GDP growth rate from 1980 to 1990 was 7.6 percent, and these years showed the greatest relative decline in Thailand’s terms of trade. The annual average GDP growth rates of the country were 4.2 percent from 1990 to 2000, 4.6 percent from 2000 to 2009, and 4.2 percent from 2009 to 2013, followed by a sharp drop to 0.7 percent in 2014. Overall, Thailand had fairly solid growth for most of the period (until 2014), and we would expect that such steady growth might well be associated, other things equal, with some decline in the terms of trade.

Finally, Jordan’s terms of trade declined the least of the four countries. There was a rather slow rise from 1980 to 1995, followed by a decline until the end of the period but with the exceptions of a sharp run-up in 2008 and 2009. Jordan’s annual average GDP growth rate was 2.5 percent from 1980 to 1990, 5.0 percent from 1990 to 2000, 7.1 percent from 2000 to 2009, 2.6 percent from 2009 to 2013, and 3.1 percent in 2014. The 2000s saw the greatest relative decline in the terms of trade (from a value of 125.4 in 1999 to 83.8 in 2014), and GDP growth was greatest from 2000 to 2009 (although there was a pronounced drop in that growth after 2009). In overview, while there is a tendency in some of the above instances for faster growth to be associated with a decline in the terms of trade (Pakistan in the early 2000s, Thailand in the 1980s, and Jordan in the early 2000s), the overall relationship of GDP to the terms of trade is a complex one. There are clearly many other factors besides GDP growth that exert important influences on the behavior of a country’s commodity terms of trade. Sources: Terms of trade calculated from data obtained at elibrary-data.imf.org; GDP growth rate data obtained from The World Bank, World Development Indicators 2002 (Washington, DC: International Bank for Reconstruction and Development/World Bank, 2002), pp. 204–6 and The World Bank, World Development Indicators 2015 (Washington, DC: International Bank for Reconstruction and Development/World Bank, 2015), pp. 82, 84–86, both available at www.worldbank.org, and from the website data.worldbank.org. Second, keep in mind that growth may lead to changes in relative demand for final products. We allowed this possibility in the discussion of the trade effects that accompany growth. In general, various classes of commodities tend to behave in a predictable way when income grows, and the different behavior patterns can be described by using the income elasticity of demand (income elasticity of demand in general, not just the income elasticity of demand for imports). For example, primary goods such as minerals and food products tend to have income elasticities less than 1.0, while manufactures tend to be characterized by an income elasticity greater than 1.0. To the extent that developing countries export labor- and land-intensive primary goods and import manufactured goods, growth in traditional export industries tends to generate protrade or ultra-protrade consumption effects that may well generate balance-of-trade deficits in fixed exchange rate economies or a depreciation of the home currency if the exchange rate is flexible. Finally, from a broader perspective, countries that rely on exports of primary goods for export earnings may find that the international prices of these goods do not rise as rapidly as the prices of the manufactured goods they import due in part to the differences in their income elasticities. This deterioration in the terms of trade certainly lowers the gains from growth in the short run and reduces the future growth rate by diminishing the ability to import needed capital goods. Economists such as Raul Prebisch (1959), Hans Singer (1950), and Gunnar Myrdal (1956) argued that the terms of trade of the developing countries declined over a long period of time, much to their disadvantage. page 224These arguments are based not only on the different demand characteristics of the two categories of products but also on the price effects of technological change. Technological advances in developing countries are assumed to lead to decreases in the prices of developing-country products, whereas in the industrialized countries, technological advances lead to increased payments to the factors of production (instead of reduced prices for manufactured goods). While it is not clear that a long-term decline in the international terms of trade of developing countries in general has taken place, it is fairly clear that there have been periods of marked short-run deterioration and improvement, often in response to unanticipated supply effects. Because primary goods tend to be less elastic than manufactures with respect to both price and income, relative price instability is also potentially a more serious problem for the developing countries than for industrialized countries. For this reason, price stabilization proposals such as commodity agreements with buffer stocks and export controls have been relatively common for developing countries (see Chapter 18).

Summary

This chapter focused on how growth in a country’s real income influences its international trade. Growth in output has an effect on a country’s trade through both consumption and production effects, which do not necessarily work in the same direction. The chapter focused on technological change and factor growth as the underlying bases for growth, and it explained the differences between the two in terms of their impact on the PPF. The effect of growth of a single factor is an expansion of production of the commodity that uses it relatively intensively and a contraction in production of the second good. The welfare effects of factor growth and technological change were positive in all small-country cases with the exception of population growth. In that case, population growth led to a fall in per capita income. The large-country case was introduced to point out the implications of growth that yields changes in the international terms of trade. Output growth in the export good generates negative terms-of-trade effects that offset some of the gains from growth. In the extreme case, a country’s welfare can decline if the effects of negative terms-of-trade change more than offset the gains from growth. Growth in production of the import-competing good can produce terms-of-trade effects that enhance the normal growth effects. Finally, this theoretical framework was used to discuss some implications of growth for the trade and development prospects of developing countries.

Chapter 12: International Factor Markets

Introduction

In this chapter, we step away from international trade in goods and services to examine the international movements of factors of production—capital and labor. The theoretical literature has long assumed that factors of production are mobile within countries, but it has also traditionally assumed that factors of production do not move between countries. This second assumption is patently false in today’s world, as we are constantly made aware of the movement of investment and labor from one country to another. We need only to note, for example, the controversies on the effect on U.S. workers of capital flows from the United States to Mexico and Asia, or the continued concern over the inflow of migrants to the United States and Europe. The constant U.S. concern about illegal immigrants from Latin America reflects the anticipated impact of large-scale labor mobility. Further, developing countries are seeking ways to restrain the outflow of skilled labor (the “brain drain”). This chapter seeks to provide an economic overview of causes and consequences of capital and labor flows. We first describe the current nature of international capital movements, discuss the principal factors that influence international investment decisions, and analyze the various effects of such investment. This is followed by a discussion of the causes and impacts of labor migration between countries.

International Capital Movements Through Foreign Direct Investment and Multinational Corporations

Foreign Investors in China: “Good” or “Bad” from the Chinese Perspective?

Few, if any, countries have ever experienced the kind of rapid economic growth that China has achieved from the end of the 1970s until the present time. World Bank data indicate that the annual average rate of increase in gross domestic product was 10.6 percent from 1990 to 2000, 10.9 percent from 2000 to 2009, and 8.7 percent from 2009 to 2013. These are growth rates that yield a doubling of GDP in every 7 to 8 years! While China’s 2013 per capita income level of $6,560 was still very low compared with that in high-income countries (e.g., per capita income in the United States in 2013 was $53,470), the growth rate was extraordinarily impressive. When allowance is made for the actual internal purchasing power of the Chinese yuan in terms of goods and services and then converting to dollars, China’s per capita income in 2013 was $11,850 rather than $6,560 and the country’s total GDP in 2013 was $16.1 trillion. This total GDP was the second largest in the world, after the $17.0 trillion GDP of the United States. (Note: The data refer to mainland China, exclusive of Taiwan and also exclusive of the separate high-income administrative region of Hong Kong.) While there have been many causes of this rapid growth, the general emphasis by economists has been placed on the liberalization of the economy that began in 1978 and featured the continuous introduction of market-oriented reforms, including greater participation in international trade. Also included in the liberalization has been the permitted entry of more foreign investors into manufacturing; such foreign direct investment (FDI) has increased dramatically. The foreign investment has been especially important in the emergence of the strong export sector—China has become the top merchandise exporting country in the world in recent years—because about one-half of Chinese exports come from firms in which foreign investors have at least some ownership share. page 228 Should China have allowed foreign investment to come into the country in such large volume? In this chapter, we analyze general economic causes and consequences of flows of capital and labor across country borders, but the Chinese case has an unusual twist that illustrates that the decision to allow foreign investment cannot be entirely economic. In an article entitled “Trade and Foreign Investment in China: A Political Economy Approach” in the December 2002 Journal of International Economics, economists Lee Branstetter and Robert Feenstra examined determinants of FDI into China during the years 1984–1995. Policies played a critical role in attracting FDI, and the policies varied by province (of which China has 30). In 1979, Guangdong and Fujian provinces on the southeast coast became sites of “special economic zones” that gave favorable tax and administrative treatment to foreign firms (more favorable treatment than to domestic Chinese firms). This favorable treatment successfully enticed foreign investors but, because the authorities did not want to endanger already-existing Chinese heavy industry, these zones were not located in China’s developed industrial areas of that time. In 1984, other areas along the coast were also permitted to give special treatment to foreign investors. In 1986, further rules permitting special tax treatment throughout China were adopted, although local regions still had regulatory powers of their own. Branstetter and Feenstra were concerned with ascertaining the factors that influenced the Chinese, by province, in their decisions regarding the allowance of greater foreign investment. In particular, the Chinese planners were hypothesized to be trading off the benefits of increased FDI (as well as increased international trade) against the losses that would be incurred by state-owned enterprises (SOEs) if foreign investment entered and, by competition against the SOEs, made the latter nonviable. To test the relevant determinants of FDI in this context, Branstetter and Feenstra looked at the provincial consumption levels of products that are provided by multinational firms who had undertaken FDI. They related the consumption levels of these FDI products to the consumption levels of similar goods produced by SOEs as well as to the levels of goods supplied as imports. An additional determinant in their testing equation was a term incorporating the wage premium paid by foreign investors, with the hypothesis being that if foreign investors pay higher wages than domestic firms, this would be an enticement for the authorities to permit more FDI so that Chinese workers would be better off. There was also a tariff revenue term, which was comprised of tariff rates (which were and still are high) times the value of imports—if tariff revenue is high, it means that potential foreign investors are supplying the Chinese market by sending in imports rather than by producing within China. What seemed to be the relationships between these various terms and production by foreign investors? The general results were that less spending on the output of Chinese state enterprises was associated with greater spending on the output of foreign investors (there was a trade-off between the two types of output), as was a higher wage premium. Higher tariff revenue collections, as expected, were associated with less foreign investor output (because foreign investors would, other things equal, be supplying from outside rather than within the country). Thus, there was a clear threat posed by FDI to production by state firms, and FDI was “bad” in that sense. Further, the fact that higher imports and consequently higher tariff revenues were associated with lower FDI meant that the government got the revenues (“good” from the state’s standpoint), but the presence of high tariffs was “bad” for consumer welfare. The higher wages paid by the foreign firms constituted “good” results from the standpoint of worker/consumer welfare. Branstetter and Feenstra then tried, in a complicated way, to integrate these results into a mathematical function that would express the government’s relative desires to promote consumer utility (by raising consumption levels and promoting higher wages), collect revenues from multinational firms (such as by imposing taxes and various fees), earn page 229profits from production by state firms, and collect tariff revenues for the government’s coffers. The most significant finding was that, although the authorities wanted to promote both state-owned production and consumer welfare, they seemed to place four to seven times as much weight on encouraging output by the SOEs as they did on promoting consumer utility. There was indeed a trading off of benefits from foreign investment against the threat of loss of viability of the state-owned production units. The politics of communism clearly played a role in this result; the populace gained in the roles of consumers and workers from having foreign investment present, but the government greatly worried that state-owned firms would take a hit from the presence of the foreign competitors. Thus, in the 1980–1995 period, China seemed to want foreign investors, but there were strong political restraining forces.

Definitions

When speaking of the international movement of “capital,” we need to distinguish two types of capital movements: foreign direct investment and foreign portfolio investment. This chapter covers FDI foreign portfolio investment is covered in international monetary economics. FDI refers to a movement of capital that involves ownership and control, as in the preceding Chinese example, where foreign ownership of production facilities took place. For example, when U.S. citizens purchase common stock in a foreign firm, say, in France, the U.S. citizens become owners and have an element of control because common stockholders have voting rights. For classification purposes, this type of purchase is recorded as FDI if the stock involves more than 10 percent of the outstanding common stock of the French firm. If a U.S. company purchases more than 50 percent of the shares outstanding, it has a controlling interest and the “French” firm becomes a foreign subsidiary. The building of a plant in Sweden by a U.S. company is also FDI, because clearly there is ownership and control of the new facility—a branch plant—by the U.S. company. FDI is usually discussed in the context of the multinational corporation (MNC), sometimes referred to as the multinational enterprise (MNE), the transnational corporation (TNC), or the transnational enterprise (TNE). These terms all refer to the same phenomenon—production is taking place in plants located in two or more countries but under the supervision and general direction of the headquarters located in one country. Foreign portfolio investment does not involve ownership or control but the flow of what economists call “financial capital” rather than “real capital.” Examples of foreign portfolio investment are the deposit of funds in a U.S. bank by a British company or the purchase of a bond (a certificate of indebtedness, not a certificate of ownership) of a Swiss company or the Swiss government by a citizen or company based in Italy. These flows of financial capital have their immediate effects on balances of payments or exchange rates rather than on production or income generation.

Some Data on Foreign Direct Investment and Multinational Corporations

The United Nations Conference on Trade and Development (UNCTAD), an organization that studies various international economic issues, has indicated that the stock of accumulated FDI inflow to countries of the world was $26,039 billion as of 2014. This $26.0 trillion stock reflected rather rapid growth in the previous decades; the stock had grown at an average annual rate of 9.4 percent from 1991 to 1995, 18.8 percent from 1996 to 2000, 13.4 percent from 2001 to 2005, 13.8 percent from 2005 to 2010 (with considerable annual variability), and 8.0 percent from 2010 to 2014. Overall, the stock of inward foreign capital of $26,039 billion in 2014 was more than 14 times as large as the stock that had been in place in 1990.2 page 230 To get a general picture of the size of FDI with respect to the United States, we present information on the amount of U.S. foreign direct investment in other countries in Table 1 and on the size of FDI in the United States in Table 2. These figures represent the total book value of accumulated FDI at the end of 2014; they are stock figures and not the flow of new investment that occurred in 2014 alone. Book value means that the numbers are basically the balance sheet figures recorded when the investments were made. Older investments are thus substantially understated relative to current value because of inflation since the time of purchase. The data indicate that the largest portion of U.S. direct investments abroad is in finance and insurance (14.4 percent) and manufacturing (13.5 percent). Geographically, European countries are the host countries (i.e., recipients) of more than one-half of U.S. FDI. Overall, the four largest recipients of U.S. direct investment in the world are the Netherlands (15.3 percent), the United Kingdom (11.9 percent), Luxembourg (9.5 percent), and Canada (7.8 percent).

Reasons for International Movement of Capital

It should be clear that there is considerable mobility of capital across country borders in the world economy today. We cannot make a full examination of the reasons for this mobility, but brief mention can be made of possible causes. Above all, economists view the movement of capital between countries as fundamentally no different from movement between regions of a country (or between industries), because the capital is moved in response to the expectation of a higher rate of return in the new location than it earned in the old location. Economic agents seek to maximize their well-being. Although many reasons for capital movements have been suggested, all imply the seeking of a higher rate of return on capital over time. We list and comment briefly on several hypotheses, many of which have found empirical support. page 233 Firms will invest abroad in response to large and rapidly growing markets for their products. Empirical studies have attempted to support this general hypothesis at the aggregative level by seeking a positive correlation between the gross domestic product (and its rate of growth) of a recipient country and the amount of FDI flowing into that country. Similarly, because manufacturing and services production in developed countries is catering increasingly to high-income tastes and wants (recall the product cycle theory from Chapter 10), it can be hypothesized that developed-country firms will invest overseas if the recipient country has a high per capita income. This suggestion leads us to expect that there would be little manufacturing investment flowing from developed countries to developing countries. However, per capita income must be kept distinct from total income (GDP), because firms in developed countries are eager to move into China because of its sheer market size and growth and despite its relatively low per capita income. Another reason for direct investment in a country is that the foreign firm can secure access to mineral or raw material deposits located there and can then process the raw materials and sell them in more finished form. Examples would be FDI in petroleum and copper. Tariffs and nontariff barriers in the host country also can induce an inflow of FDI. If trade restrictions make it difficult for the foreign firm to sell in the host-country market, then an alternative strategy for the firm is to “get behind the tariff wall” and produce within the host country itself. It has been argued that U.S. companies built such tariff factories in Europe in the 1960s shortly after the European Economic Community (Common Market) was formed, with its common external tariff on imports from the outside world. Such U.S. investment continued in the 1990s as Europe pressed for even closer economic integration and adopted a common currency for 11 countries in 1999 (now 19 countries). A foreign firm may consider investment in a host country if there are low relative wages in the host country, although studies indicate that low wages per se are not as much an enticement for FDI as envisioned by the general public. Clearly, the existence of low wages because of relative labor abundance in the recipient country is an attraction when the production process is labor intensive. In fact, the production process often can be broken up so that capital-intensive or technology-intensive production of components takes place within developed countries while labor-intensive assembly operations that use the components take place in developing countries. This division of labor is facilitated by offshore assembly provisions in the tariff schedules of developed countries (see Chapter 13). Firms also argue that they need to invest abroad to protect foreign market share. Firm A, for instance reasons that it needs to begin production in the foreign market location in order to preserve its competitive position because its competitors are establishing plants in the foreign market currently served by A’s exports or because firms in the host country are producing in larger volume and competing with A’s goods. A recent example is Toyota Motors, which completed building production facilities abroad because the high value of the yen had reduced its competitiveness in foreign markets.3page 234 It has also been suggested that firms may want to invest abroad as a means of risk diversification. Just as investors prefer to have a diversified financial portfolio instead of holding their assets in the stock of a single company, so firms may wish to distribute their real investment assets across industries or countries. If a recession or downturn occurs in one market or industry, it will be beneficial for a firm not to have all its eggs in one basket. Some of the firm’s investments in other industries or countries may not experience the downturn or may at least experience it with reduced severity. Foreign firms may find investment in a host country to be profitable because of some firm-specific knowledge or assets that enable the foreign firm to outperform the host country’s domestic firms (see Graham and Krugman, 1995, chap. 2; and Markusen, 1995). Superior management skills or an important patent might be involved. At any rate, the opportunity to generate a profit by exploiting this advantage in a new setting entices the foreign firm to make the investment.

Determinants of Foreign Direct Investment

Numerous econometric studies have attempted to ascertain the factors that cause FDI flows between countries. Reinhilde Veugelers (1991) examined data for 1980 on FDI from developed countries to other developed countries to determine why some recipient countries were chosen over others. The dependent variable in Veugelers’s regression analysis was the number of foreign affiliates (plants abroad with at least some home firm control) of any country i located in recipient country j as a percentage of the total foreign affiliates of country i. With respect to the independent variables, a statistically significant positive relationship was found with the GDP of the recipient country, weighted by the degree of openness of the recipient. This finding reflects the importance of market size and possible economies of scale. The weight for openness was included in recognition of the engagement of foreign affiliates in export and in recognition that a recipient country’s greater openness to trade would permit greater exports from any affiliate. Veugelers also found a positive relationship with FDI when the sending and receiving countries shared a common language or common boundaries. However, a negative relationship was found with the ratio of fixed investment to GDP in the recipient country; this was surprising because Veugelers had expected that a high fixed-investment ratio would mean a relatively large amount of infrastructure and thus an inducement for foreign investors. Finally, labor productivity in the recipient country, distance between the sending and receiving countries, and tariff rates in both sets of countries had insignificant impacts. In an earlier study, Franklin Root and Ahmed Ahmed (1979) examined possible influences on the inflow of FDI into the manufacturing sector in a sample of 58 developing countries. Six variables seemed to be most important. Other things being equal, the amount of FDI was greater: (a) the higher the per capita GDP of the host country; (b) the greater the growth rate in total GDP of the host country; (c) the greater the degree of recipient country participation in economic integration projects such as customs unions and free-trade areas; (d) the greater the availability of infrastructure facilities (e.g., transport and communication networks) in the recipient country; (e) the greater the extent of urbanization of the recipient country; and (f) the greater the degree of political stability in the host country. A later study by Ray Barrell and Nigel Pain (1996) examined possible determinants of U.S. direct investment abroad during the 1970s and 1980s. In their econometric work, they found that world market size (as measured by the combined GNPs of the seven largest industrialized countries) was a stimulant to U.S. FDI, with a 1 percent rise in the combined GNPs leading to an increase of 0.83 percent in the stock of U.S. investment facilities abroad. In addition, they found a positive relationship between U.S. FDI and the level of U.S. labor costs relative to labor costs in Canada, Japan, Germany, France, and the United Kingdom. The statistical estimate page 235was that an increase of 1 percent in relative U.S. labor costs raised U.S. FDI by 0.49 percent. A positive association was also evident between U.S. FDI and U.S. relative capital costs. Further, there was some positive relation between U.S. FDI and domestic profits in the United States—suggesting an “availability of funds” cause. Besides these findings regarding the role of market size, relative labor and capital costs, and profits, an interesting result pertained to the exchange rate. An expected rise in the value of the dollar relative to other currencies led to some temporary postponement of U.S. foreign direct investment, suggesting that payments abroad associated with making the investment are delayed in anticipation of the greater command over foreign currencies that the dollar will have when the appreciation eventually takes place. A 2002 paper by Romita Biswas examined econometrically the determinants of U.S. foreign direct investment in 44 countries from 1983 to 1990. In particular, Biswas focused on the influence of compensation paid per employee, infrastructure in the receiving country (with infrastructure being measured by installed capacity of electric generating plants per capita and by the number of main telephone lines per capita), and total GNP. Further, political variables such as type of regime in place (autocracy or democracy), regime duration, rule of law, property rights (such as extent of protection from expropriation by the government), and amount of corruption in government were also included in the empirical analysis. (Obviously, some of these variables would be difficult to measure!) In general, infrastructure was found to contribute positively and significantly to the receipt of FDI, higher wages meant less FDI (although not in all tests), and democracies were more attractive to FDI than were autocracies. Greater protection of property rights also enhanced FDI. Curiously, a longer duration of a regime significantly reduced FDI. Biswas hypothesizes that this result might have occurred because the longer a regime is in place, the greater the chance that interest groups will form—groups that decrease the flexibility and efficiency of government. Finally, an interesting paper by Judith Dean, Mary Lovely, and Hua Wang (2009) addressed the question of whether environmental regulations have an impact on incoming FDI. A standard hypothesis is that firms in high-income countries will, other things equal, tend to locate their production facilities in low-income countries rather than in their own home countries because of the stricter environmental standards in place in the high-income countries (often called a “race to the bottom” with respect to environmental protection). Dean, Lovely, and Wang focused on China in the 1990s, using a data set that contained almost 3,000 FDI joint-venture manufacturing facilities. (The joint-venture enterprise involves combined ownership by the foreign investor and a host country firm/government and was the common type of FDI in China during the time period.) Because environmental standards differed across provinces in China, the study attempted to determine whether these different standards, after allowing for other influences on FDI, were a factor in foreign investors’ choosing to locate in low-standard provinces rather than in high-standard provinces. Environmental regulations were represented by the Chinese water pollution levy system, in which firms faced a tax if certain types of pollutants were discharged or if specified volume and concentration levels of pollution were exceeded. The tax rate varied considerably across the provinces. In the paper, the authors concluded that FDI in high-pollution industries from ethnically Chinese sources (which included Hong Kong, Macao, and Taiwan) was significantly deterred from going into the provinces with the higher environmental standards. However, the provincial location of FDI from origins that were not ethnically Chinese did not appear to be affected by the differing levels of environmental regulation. In overview, there are clearly many different possible factors leading to FDI. Important attention is being paid, and rightfully so, to noneconomic variables as well as to traditional economic variables. Finally, differences in government policies can influence rates of return and therefore the location of firms. For example, a contentious issue in the United States in recent years is that some domestic firms have made the decision to merge with a foreign firm and locate the headquarters of the combined firms abroad in response to the relatively more favorable tax treatment in the foreign location. This phenomenon is known as “inversion.” A considerable amount of further empirical research is needed to determine the most important causes of international capital mobility, and different reasons will apply to different industries, different periods, and different investors.

The Effects of International Capital Movements

A Theoretical Framework for Analyzing International Capital Movements

The existence of substantial international capital mobility in the real world has various implications for the output of the countries involved, for world output, and for rates of return to capital and other factors of production. Economists employ a straightforward microeconomic apparatus to examine these effects, and this section presents this analytical approach. We return to this apparatus in our discussion of the international movement of labor later in the chapter. Figure 1 portrays the marginal physical product of capital (MPPK) schedules for countries I and II. The analysis assumes that they are the only two countries in the world, that there are only two factors of production—capital and labor—and that both countries produce a single, homogeneous good that represents the aggregate of all goods produced in the countries. In microeconomic theory, a marginal physical product of capital schedule plots the additions to output that result from adding 1 more unit of capital to production when all other inputs are held constant. With constant prices, this schedule constitutes the demand for capital inputs derived from the demand for the product. Schedule AB shows the MPPK in country I (MPPK I) for various levels of capital stock measured in a rightward direction from origin 0. Analogously, schedule A′B′ indicates the MPPK in country II (MPPK II), with the levels of capital stock measured leftward from origin 0′. Assume in the initial (pre-international-capital-flow) situation that the capital stock in country I is measured by the distance 0k1 and capital in country II is measured (in the leftward direction) by the distance 0′k1. The total world capital stock is fixed and equal to the distance 00′, or the sum of 0k1 and 0′k1. With the standard assumption of perfect competition, capital in country I will be paid at the rate equal to its marginal product (0r1), which is associated with point C on schedule AB. Similarly, capital in country II will be paid at the rate equal to its marginal product (0′r′1), which is associated with point C′ on schedule A′B′. Remembering that total product is equal to the area under the marginal product curve at the relevant size of capital stock, the total output (or GDP) in country I is equal to area 0ACk1 and the total output (GDP) in country II is equal to area 0′A′C′k1. (World output is of course equal to the sum of these two areas.) The total output in country I is divided between the two factors such that the rectangle 0r1Ck1 is the total return (or profit) of capital (i.e., the rate of return 0r1 multiplied by the amount of capital 0k1), and workers receive the remaining output (or income) consisting of triangle r1AC. In country II, by similar reasoning, capital receives total return (or profit) of area 0′r′1C′k1 and labor receives the area of triangle r′1A′C′.

Host-Country Determinants of Foreign Direct Investment Inflows

The United Nations Conference on Trade and Development (UNCTAD), in its World Investment Report 1998, categorized types of FDI and the general characteristics of host countries that are considered by investors deciding whether to undertake a project in any given country. These factors have also been elaborated on in the context of developing countries in a 1999 article in Finance and Development (Mallampally and Sauvant, 1999). The particular economic determinants of FDI, according to the UNCTAD staff, depend on whether the FDI project falls into one of three categories: (1) market-seeking FDI, that is, firms that are attempting to locate facilities near large markets for their goods and services; (2) resource-seeking and asset-seeking FDI, that is, firms that are in search of particular natural resources (e.g., copper in Chile) or particular human skills (e.g., computer literacy and skills in Bangalore, a city in southern India often referred to as the “Second Silicon Valley”); and (3) efficiency-seeking FDI, that is, firms that can sell their products worldwide and are in search of the location where production costs are the lowest. These general economic determinants are listed in the left-hand column of Table 5. Beyond economic factors, foreign firms considering investment in any given country will also be influenced by various policies and attitudes of the host country’s government. In addition, broader, more general characteristics of the business environment (called “business facilitation” by UNCTAD) will play a role in the investment decision. These policy and business environment considerations, as presented by UNCTAD, are listed in the right-hand column of Table 5. In general, the table gives us a framework for viewing the decision to undertake FDI in any given case. Of course, the weights to be applied to each factor will differ from potential host country to potential host country, and different weights will also be applied by different foreign firms.

Source: Padma Mallampally and Karl P. Sauvant, “Foreign Direct Investment in Developing Countries,” Finance and Development 36, no. 1 (March 1999), p. 36. Originally appeared in United Nations Conference on Trade and Development, World Investment Report 1998: Trends and Determinants (Geneva: UNCTAD, 1998), p. 91.

This situation will change if capital is permitted to move between countries because the rate of return to capital in country I (0r1) exceeds that in country II (0′r′1). If capital mobility exists between the two countries, then capital will move from country II to country I as long as the return to capital is greater in country I than in country II. (We are assuming that the same degree of risk attaches to investments in each country or that the rates of return have been adjusted for risk. We are also assuming that there is no international movement of labor.) In Figure 1, the amount of capital k2k1 in country II moves to country I to take advantage of the higher rate of return. This FDI by country II in country I bids down the rate of return in country I to 0r2. On the other hand, because capital is leaving country II, the rate of return in country II rises from 0′r′1 to 0′r′2. In equilibrium, the MPPK in the two countries is equal, and this is represented by point E, where the two marginal physical product of capital schedules intersect. At this equilibrium, the rate of return to capital is equalized between the countries (at 0r2 = 0′r′2), and there is no further incentive for capital to move between the countries. What has been the effect of capital flow k2k1 from country II to country I on output in the two countries and on total world output? As expected, total output has risen in country I because additional capital has come into the country to be used in the production process. Before the capital flow, output in country I was area 0ACk1, but output has now increased to area 0AEk2. Thus, output in country I has gone up by the area k1CEk2. In country II, there has been a decline in output. The before-capital-flow output of 0′A′C′k1 has been reduced to the after-capital-flow output of 0′A′Ek2, a decrease by the amount k1C′Ek2. However, world output and thus efficiency of world resource use has increased because of the free movement of capital. World output has increased because the increase in output in country I (area k1CEk2) is greater than the decrease in output in country II (area k1C′Ek2). The extent to which world output has increased is indicated by the triangular shaded area C′CE. Thus, just as free international trade in goods and services increases the efficiency of resource use in the world economy, so does the free movement of capital—and of factors of production in general. In addition, free movement of factors can equalize returns to factors in the two countries, just as free international trade in the Heckscher-Ohlin model could lead to factor price equalization between the countries. In recognition of these parallel implications of trade and factor mobility for efficiency of resource use and returns to factors, economists often stress that free trade and free factor mobility are substitutes for each other. Some comments also can be made about the total return to each of the factors of production in the two countries. The total return to country I’s owners of capital was 0r1Ck1 before the capital movement, but it has now fallen to the amount 0r2Fk1 (a decline by the amount r2r1CF). The return to country II’s owners of capital has increased from 0′r′1C′k1 to 0′r′2Fk1, an increase by the amount r′1r′2FC′. While we know that owners of capital in country I have been injured and those in country II have gained from the capital flow, we cannot say anything about the sum of the two returns (and thus of world profits) unless more information page 239is available on the slopes of the MPPK schedules and the size of the capital flow. However, because world output has increased, it is theoretically possible to redistribute income so that both sets of capital owners could be better off than they were prior to the capital movement. A similar conclusion applies to labor. Workers in country I have received an increase in their total wages, because before-capital-flow wages consisted of area r1AC while after-capital-flow wages are indicated by area r2AE (an increase in wages by the amount r2r1CE). In country II, wages have fallen because workers now have less capital with which to work. The wage bill in country II prior to the capital flow was area r′1A′C′, and it has decreased to r′2A′E after the capital flow(a decrease by the amount r′1r′2EC′). Again, no a priori statement can be made about the impact of the capital flow on total wages in the world without more information, but the increase in world output (and income) suggests that all workers could be made better off by income redistribution policies. Finally, we can make unambiguous statements about the impact of the capital flow on national income [or gross national product (GNP)—the product of a country’s nationals or citizens] in both countries. The income of country I’s citizens consists of total wages plus total profits. We have seen that the capital flow has increased total wages by area r2r1CE and has decreased the returns to the owners of capital by area r2r1CF. Comparison of these two areas indicates that the income of workers rises by more than the income of capital owners falls in country I; we conclude that national income or GNP—the income of the factors of production—in country I increases because of the capital inflow (by triangular area FCE). (GDP—the total output produced within a country—for country I has risen by k1CEk2. However, area k1FEk2 of that amount accrues to country II’s investors.) Analogously, the capital outflow in country II causes total wages to fall by area r′1r′2EC′ and the total returns to owners of capital to rise by area r′1r′2FC′. National income (GNP) in country II thus increases by amount C′FE. Country II has higher income (GNP) despite the fact that the output produced in II (its GDP) has fallen from area 0′A′C′k1 to area 0′A′Ek2. Hence, both countries gain from international capital mobility. Restrictions on the flow of FDI have an economic cost of lost efficiency in the world economy and lost income in each of the countries.

Potential Benefits and Costs of Foreign Direct Investment to a Host Country

In this section, we cover some of the alleged benefits and costs of a direct capital inflow to a host country. (For an expanded discussion of many of these points, see Meier, 1968, 1995.) While there are also benefits and costs to the home country from capital outflow, we focus only on host-country effects. The focus on impacts to the host country particularly permits us to discuss developing countries more prominently.

Potential Benefits of Foreign Direct Investment

A wide variety of benefits may result from an inflow of FDI. These gains do not occur in all cases, nor do they occur in the same magnitude. Several of the potential gains are listed here. Increased output. This impact was discussed earlier. The provision of increased capital to work with labor and other resources can enhance the total output (as well as output per unit of input) flowing from the factors of production. Increased wages. This was also discussed earlier. Note that some of the increase in wages arises as a redistribution from the profits of domestic capital. Increased employment. This impact is particularly important if the recipient country is a developing country with an excess supply of labor caused by population pressure. Increased exports. If the foreign capital produces goods with export potential, the host country is in a position to generate scarce foreign exchange. In a development context, the additional foreign currency can be used to import needed capital equipment or materials to assist in achieving the country’s development plan, or the foreign exchange can be used to pay interest or repay some principal on the country’s external debt.page 240 Increased tax revenues. If the host country is in a position to implement effective tax measures, the profits and other increased incomes flowing from the foreign investment project can provide a source of new tax revenue to be used for development projects. However, the country must spend such revenue wisely and refrain from imposing too high a rate of taxation on the foreign firm, as this high taxation might cause the firm to leave the country. Realization of scale economies. The foreign firm might enter into an industry in which scale economies can be realized because of the industry’s market size and technological features. Home firms might not be able to generate the necessary capital to achieve the cost reductions associated with large-scale production. If the foreign investor’s activities realize economies of scale, consumer prices might be lowered. Provision of technical and managerial skills and of new technology. Many economists judge that these skills are among the scarcest resources in developing countries. If so, then a crucial bottleneck is broken when foreign capital brings in critical human capital skills in the form of managers and technicians. In addition, the new technology can clearly enhance the recipient country’s production possibilities. Weakening of power of domestic monopoly. This situation could result if, prior to the foreign capital inflow, a domestic firm or a small number of firms dominated a particular industry in the host country. With the inflow of the direct investment, a new competitor is provided, resulting in a possible increase in output and fall in prices in the industry. Thus, international capital mobility can operate as a form of antitrust policy. A recent example of the potential for this is the effort by U.S. telecommunications firms to gain greater access to the Japanese market. The difficulties associated with competition from foreign investors were illustrated in 2011 when the Indian government decided to permit foreign investors in the retail sector to form joint ventures with local retailers whereby the foreign investor could have majority ownership. Domestic protests resulted in the government rescinding this decision. The only foreign-investor-dominated joint ventures then permitted were for single-brand retailers (e.g., Starbucks, Nike). However, it was decided in 2012 that many-brand retailers (e.g., Walmart, Target) could also enter.

Potential Costs of Foreign Direct Investment

Some alleged disadvantages to the host country from a foreign capital inflow are listed and briefly discussed. Adverse impact on the host country’s commodity terms of trade. As you will recall, a country’s commodity terms of trade are defined as the price of a country’s exports divided by the price of its imports. In the context of FDI, the allegation is sometimes made that the terms of trade will deteriorate because of the inflow of foreign capital. This could occur if the investment goes into production of export goods and the country is a large country in the sale of its exports. Thus, increased exports drive down the price of exports relative to the price of imports. Transfer pricing is another mechanism by which the host country’s terms of trade could deteriorate. The term transfer prices refers to the recorded prices on intrafirm international trade. If one subsidiary or branch plant of a multinational company sells inputs to another subsidiary or branch plant of the same firm in another country, no market price exists; the firm arbitrarily records a price for the transaction on the books of the two subsidiaries, leaving room for manipulation of the prices. If a subsidiary in a developing country is prevented from sending profits home directly or is subject to high taxes on its profits, then the subsidiary can reduce its recorded profits in the developing country by understating the value of its exports to other subsidiaries in other countries and by overstating the value of its imports from other subsidiaries. What happens is that the country’s recorded terms of trade are worse than they would have been if a true market price was used for these transactions. page 241 Decreased domestic saving. The allegation, in the context of a developing country, is that the inflow of foreign capital may cause the domestic government to relax its efforts to generate greater domestic saving. If tax mechanisms are difficult to put into place, the local government may decide there is no need to collect more taxes from a low-income population for the financing of investment projects if a foreign firm is providing investment capital. The forgone tax revenues can be used for consumption rather than saving. This is only one of several possible mechanisms for achieving the same result. Decreased domestic investment. Often the foreign firm may partly finance the direct investment by borrowing funds in the host country’s capital market. This action can drive up interest rates in the host country and lead to a decline in domestic investment through a “crowding-out” effect. In a related argument, suppliers of funds in the developing country may provide financial capital to the MNC rather than to local enterprise because of perceived lower risk. This shift of funds may divert capital from uses that could be more valuable to the developing countries. Instability in the balance of payments and the exchange rate. When the FDI comes into the country, it usually provides foreign exchange, thus improving the balance of payments or raising the value of the host country’s currency in exchange markets. However, when imported inputs need to be obtained or when profits are sent home to the country originating the investment, a strain is placed on the host country’s balance of payments and the home currency can then depreciate in value. A certain degree of instability will exist that makes it difficult to engage in long-term economic planning. Loss of control over domestic policy. This is probably the most emotional of the various charges levied against FDI. The argument is that a large foreign investment sector can exert enough power in a variety of ways so that the host country is no longer truly sovereign. For example, this charge was levied forcefully against U.S. direct investment in western Europe in the 1960s and it has often been raised against U.S. FDI into developing countries. Also, the U.S. government has in place a Committee on Foreign Investment in the United States (CFIUS) that examines proposed FDI projects in the United States with respect to their impact on national security. If security is likely to be endangered, the FDI will not be permitted. Increased unemployment. This argument is usually made in the context of developing countries. The foreign firm may bring its own capital-intensive techniques into the host country; however, these techniques may be inappropriate for a labor-abundant country. The result is that the foreign firm hires relatively few workers and displaces many others because it drives local firms out of business. Establishment of local monopoly. This is the converse of the presumed “benefit” that FDI would break up a local monopoly. On the “cost” side, a large foreign firm may undercut a competitive local industry because of some particular advantage (such as in technology) and drive domestic firms from the industry. Then the foreign firm will exist as a monopolist, with all the accompanying disadvantages of a monopoly. Inadequate attention to the development of local education and skills. First propounded by Stephen Hymer (1972), this argument has the multinational company reserving the jobs that require expertise and entrepreneurial skills for the head office in the home country. Jobs at the subsidiary operations in the host country are at lower levels of skill and ability (e.g., routine management operations rather than creative decision making). The labor force and the managers in the host country do not acquire new skills.

Overview of Benefits and Costs of Foreign Direct Investment

No general assessment can be made regarding whether the benefits outweigh the costs. Each country’s situation and each firm’s investment must be examined in light of these various considerations, and a judgment about the desirability of the investment can be clearly positive in some instances and negative in others. These considerations get us page 242beyond the simple analytical model discussed earlier in this chapter, where the capital flow was always beneficial in its impact. Developed and developing countries often try to institute policies that will improve the ratio of benefits to costs connected with a foreign capital inflow. Thus, performance requirements are frequently placed on the foreign firm, such as stipulating a minimum percent of local employees, a maximum percent of profits that can be repatriated to the home country, and a minimum percent of output that must be exported to earn scarce foreign exchange. In addition, the output of the firm may be subject to domestic content requirements on inputs, or foreign firms may be banned altogether from certain key industries. Some progress toward eliminating such distortionary performance requirements was made in the Uruguay round of trade negotiations in the 1990s. Finally, brief mention can be made of the fact that clearly there are impacts of FDI on the sending or home country of the investment as well as on the receiving or host country. As noted in the discussion of Figure 1, the sending country (country II in the figure) experiences a reduction in its GDP (although an increase in its national income or gross national product), a reduction in total wages, and an increase in the total return to its investors. The country could also undergo such effects as a loss of tax revenue from the investing firms (depending on tax treaty arrangements between the sending and the receiving country of the FDI) and a loss of jobs. International trade could also be affected—for example, exports from the FDI-sending country could rise if the new plants abroad obtained inputs from home sources. Alternatively, exports from the sending country could fall if the new plant was set up abroad to supply the foreign market from the foreign country itself rather than through export from the home country (as in the product cycle theory in Chapter 10). On the import side, imports into the home country could increase if the new FDI plant assembles or produces relatively labor-intensive products in a relatively labor-abundant host country and the home country is a relatively capital-abundant country. Other effects in practice, of course, depend on the particular investment project being considered.

Labour Movements between Countries

Seasonal Workers in Germany

The Winkelmann farming group, headed by two brothers, grew, in a relatively short time, from a local asparagus farm in Germany to a position as one of the top 10 white asparagus suppliers in the country. The firm relies heavily on temporary immigrant workers for harvesting its crops—from a situation of owning 2.5 acres and using two migrant workers in 1989, the Winkelmanns expanded into the former Democratic Republic of Germany (East Germany) after German reunification in 1990 and, in 2002, owned 2,500 acres of land and employed almost 4,000 migrant workers. Workers, 80 percent of whom are Polish, are hired after a thorough recruitment process page 243that includes extensive background checks and training in the workers’ home country. The workers are employed for three months per year, and they are then sent home, with transportation for the trip home paid for by the Winkelmanns. (The Winkelmanns employ only workers who have a job at home, a job to which they can return after the three months’ employment in Germany has been completed.) While in Germany, the temporary migrants receive housing and insurance from the Winkelmanns, and Polish workers can earn wages in the three months that are equivalent to 150 percent of a year’s pay in Poland. This temporary migration system is of considerable value to the Winkelmanns and to other farms like theirs, but it also appears to benefit Germany in its agricultural production. Germany gains because it has been difficult to recruit Germans to harvest the asparagus, apparently because the work is physically demanding and pays relatively low wages (relatively low for the Germans but not for the Poles).

Permanent Migration: A Greek in Germany

Hasan Touzlatzi is a Muslim from West Thrace, Greece, who lives in Espelkamp, a small town in Germany. He grew up in a poor family in Greece, and he left West Thrace at age 20 in 1970 to go to Germany for temporary work. Hasan traveled to Germany with other temporary “guest workers,” and the trip had been organized by the German government. He was provided with a job in a firm in Espelkamp, and, at least partly because he began learning the German language as soon as he arrived in the country, he advanced quickly with the firm. When the firm later folded, Hasan decided on several successive occasions, although planning only for a short extension on each occasion, to stay on in his new country. His wife joined him and, after children were born, the Touzlatzis became permanent residents so that their children could benefit from the German education system. Hasan Touzlatzi has become a respected and prominent member of the Espelkamp community, where he has lived for more than 30 years. He owns a flower shop, is active in a local club of immigrants from West Thrace, and participates regularly in the Espelkamp Muslim prayer room and mosque. He and his family and other fellow migrants are solid parts of the German community and economy, although ties continue with their homeland. (For example, two of Hasan’s sons went to Greece to serve in the Greek army, and Hasan has kept his Greek citizenship.) The Touzlatzis are permanent immigrants who have become integrated into their host country, although they retain identification with their homeland. These two vignettes offer examples of temporary migration and permanent migration between countries. Just as capital moves in large volume across country borders, so too does labor. The World Bank has estimated that, in 2013, 247 million people, or about 3.4 percent of the world population did not reside in the countries in which they were born.5 On an individual country basis, as examples, 23.9 percent of Australia’s population were foreign-born in 2006, 9.7 percent of the United Kingdom’s population in 2005 had been born in another country, and, for Spain, the figure was 13.1 percent in 2008.6 For the United States in 2012, 40.0 million people were foreign-born,7 which constituted 12.9 percent of the population. In addition, of course, there has been, over the past few decades, considerable illegal as well as legal migration into the United States, with the illegal immigration (generally thought to be about 11 million people) having been extremely controversial. This number is lower than other estimates, but it is also likely that the size of the illegal immigrant population declined in several years immediately following 2007 and 2008, when recession conditions dampened the job prospects for immigrants and led to a reduced infloaw. While there are many different reasons for large-scale migration, including economic, political, and familial ones, we focus mainly on the economic causes and consequences in this chapter.

Migration Flows into the United States, 1986 and 2013

As is well known and has been the source of considerable controversy, the number of annual migrants into the United States has increased in recent decades. Table 6 gives data on the total number of immigrants and their sources* for 1986 and 2013. In 1986 there were 601,708 immigrants; in 2006 the annual inflow had more than doubled to 1,266,129, but declined by 2013 to 990,553. (Note, of course, that there is likely to be some understatement in the totals because it is very difficult to get a precise count of all immigrants.) Beyond these totals, Table 6 also indicates the regions of origin of the migrants, as well as the leading countries of origin. As can be seen, the two largest regional sources are the Americas and Asia. Asian immigrants in 1986 constituted 44.6 percent of the total flow, while migrants from Latin America and the Caribbean accounted for 41.5 percent. These two regions continued to dominate in 2013, with the number of immigrants from the Americas falling slightly to 40.0 percent and those from Asia falling to 40.4 percent. Whereas the large majority of U.S. immigrants in the late 19th and early 20th centuries came from Europe, the European countries sent only 10.4 percent of the total U.S. immigrants in 1986 and only 8.7 percent in 2013. Looking at the countries of origin, Mexico was the leading source country in both years, with 11.1 percent in 1986 and 13.6 percent in 2013. The absolute number of Mexican immigrants in 2013 was more than twice the number of Mexican immigrants in 1986, although declining in number in very recent years. China was the second-largest source country in 2013 (7.2 percent of the total), while it had been eighth-largest in 1986. The number of Chinese immigrants in the annual flow nearly tripled between the two years. The Philippines, which had been the second-largest source in 1986, was fourth-largest in 2013. India sent 68,458 migrants (third-largest source at 6.9 percent) in 2013 (many of them entering the United States under the H-1 skilled-labor visa program), compared with 26,227 in 1986 (sixth-largest source at 4.4 percent). Finally, the Republic of Korea, which had been the third-largest source country in 1986, dropped to eighth-largest in 2013. *Inflow of new legal permanent residents by country of birth.

Technically, the desire to migrate on the part of an individual depends on the expected costs and benefits of the move. Expected income differences between the old and new location, costs of the move, cost-of-living differences between the two locations, and other nonpecuniary net benefits in the new location such as health facilities, educational opportunities, or greater political or religious freedom figure into the decision to migrate. Even within this more general framework, expected wage or income differences are an important factor. At the same time, the movement of labor can influence the average wage in both the old and the new locations. For both countries, the movement of labor thus has welfare implications similar to capital movements and trade in goods and services.

Economic Effects of Labor Movements

The economic implications of labor movements between countries can be observed most readily by using a figure similar to that used for capital. Assuming that labor is homogeneous in the two countries and mobile, labor should move from areas of abundance and lower wages to areas of scarcity and higher wages. This movement of labor causes the wage rate to rise in the area of out-migration and to fall in the area of in-migration. In the absence of moving costs, labor continues to move until the wage rate is equalized between the two regions (see Figure 2). The labor force of both countries is represented by the length of the horizontal axis. The demand (the marginal physical product) for labor in each country is denoted by demand curves DI and DII. If markets are working perfectly and labor is mobile, the wage in both countries should settle at 0Weq, and 0L1 labor will be employed in country I and L10′ in country II. Suppose that the markets have not jointly cleared and that the wage in country I remains below that of country II. This would be the result if 0L2 existed in country I and country II had only L20′ labor. If labor now responds to the wage difference, labor should move from country I to country II. As this takes place, the wage in country I should rise while that in country II should fall until 0Weq exists in both countries. As these adjustments occur, output falls in country I and rises in country II. The remaining laborers in country I are better off both absolutely (due to the higher wage) and relatively, as the productivity of the other factors falls with the reduced labor supply. In country II, the opposite takes place. With the fall in the wage rate in country II, labor is less well-off. Productivity of the other factors, however, has risen with the increased use of labor, so owners of these factors are better off. The other factors in country II gain area ABFGD, while country II’s labor loses area DBFG. The amount of income earned by the new migrants is L1ADL2.

What can be said about the change in overall well-being in country I, country II, and the world as a result of the labor movement? Given the existence of diminishing marginal productivity of labor in production, other things being equal, output (GDP) in country I falls at a slower rate than the decrease in the labor force, leading to an increase in per capita output. In country II, output (GDP) grows more slowly than the increase in the labor force, leading to a decrease in per capita output. Finally, the world as a whole gains from this migration since the fall in total output in country I (area L1ACL2) is more than offset by the increase in output in country II (area L1ABL2) by the shaded area ABC. An even clearer case of world gains from migration occurs if it is assumed that market imperfections within country I lead to an initial excess supply of labor. Now not only do wages differ between country I and country II, but also some labor remains unemployed in country I at the institutional (traditional) wage rate. This above-equilibrium wage could be the result of minimum wage laws and labor union-induced downward wage rigidity in manufacturing or of the existence of an agricultural sector where families simply divide up farm output among all members (workers thus receive their average product, not their marginal product). This excess supply is often called surplus labor in the economic development literature. Figure 3 shows distance L20′ as the amount of labor available in country II, and distance 0L2 as the amount of labor in country I. The labor in country II is employed at the domestic equilibrium wage of 0′WII while in country I the prevailing wage rate is 0WI (instead of the lower, market-clearing 0W′Ieq), leading to only 0L1 people being employed. L1L2 people are thus currently unemployed at the prevailing wage rate. Migration of unemployed workers L1L2 from country I to country II in this case leads to an expansion of output in country II without any reduction in output in country I. Complete equalization of wages requires that additional LeqL1 workers move from country I to country II so that Leq0′ workers are employed in country II. If this additional migration occurs, output in country I declines because previously employed labor, LeqL1, leaves the country. The effect of migration resulting from surplus labor, while similar in direction to that in the earlier full-employment case, produces different magnitudes of results. The gain in per capita output in country I caused by the migration is clearly greater because the loss of unemployed workers, L1L2, does not affect country I’s total output. The increase in total output and the decline in per capita output in country II is the same as before (see Figure 2), and the net world gain (area ABC plus area L1DCL2—the shaded area) is larger by L1DCL2, that is, the value of production forgone in country I as a result of the unemployment. This example points out that the greater the number of market imperfections—in this case a domestic market distortion (failure of the domestic labor market in country I to clear) and an international distortion (differential wage rates across countries)—the greater the potential gains from removing these distortions.

Migration of labor (or capital) also affects the composition of output and structure of trade in the countries involved. The inflow of labor into country II is similar in effect to growth in the labor force discussed in Chapter 11 (see Figure 4). Given full employment, at constant international prices, the increase in the labor force in country II leads, according to the Rybczynski theorem, to an expansion of output of the labor-intensive good (textiles) and a contraction in output of the capital-intensive good (autos). Assuming that country I is the labor-abundant country, that country II is the capital-abundant country, and that trade between the two follows the Heckscher-Ohlin pattern, the effect of the labor movement between the two can be examined. Output of the export good in country II declines and output of the import good increases. Thus, the production trade effect is an ultra-antitrade effect.

The movement of labor from country I to country II is indicated by the outward shift of the PPF for country II and the inward shift of the PPF for country I. Assume that country I is the labor-abundant country exporting the labor-intensive good (textiles) and importing the capital-intensive good (autos) prior to the labor migration and that the two countries in question are small countries. The Rybczynski theorem indicates that this change in relative labor supplies will lead country I to contract production of textiles (the labor-intensive good) from t0 to t1 and expand production of autos from a0 to a1. Country II, on the other hand, will expand production of textiles from T0 to T1 with the newly acquired labor and reduce the production of autos from A0 to A1. Both production adjustments are ultra-antitrade in nature since factor flows have in effect substituted for trade flows. page 248 In a similar fashion, the reduction in labor in country I causes production of the labor-intensive good to fall and production of the capital-intensive good to rise. The production effects in both countries are symmetric and are ultra-antitrade in nature. The total effect of the labor movement on the volume and structure of trade will ultimately depend not only on the production effects but also on the consumption effects, which reflect the income changes and the income elasticity of demand for the two products in both countries. Finally, this analysis assumes the absence of any price distortions in either country and assumes that international prices do not change as a result of the factor movements. Price distortions and changes in international prices could alter these conclusions. The analysis of factor movements with price distortions and world price changes is beyond the scope of this text.

Additional Considerations Pertaining to International Migration

The previous models help us understand some of the basic issues that affect the politics of labor migration. It is not surprising that labor in country II wants restrictions against immigration because new workers lower the wage rate. For example, in early 2009, strikes occurred in the United Kingdom as workers protested that the French oil firm Total had awarded a U.K. construction contract to a company that would bring in foreign workers for use in production in the United Kingdom.8 On the other hand, owners of other resources such as capital favor immigration because it increases their returns. At the same time, labor in country I favors out-migration (emigration), while capital owners tend to discourage the labor movement. While the simple models are useful in providing an understanding of the basic economics involved, several extensions of this analysis are important to discuss briefly. First, the new immigrant might transfer some income back to the home country. When this happens, the reduction in income (from home production) in country I is at least partly offset by the amount of the transfer, while the increase in income resulting from the increased employment in country II is reduced by the amount of the transfer. Assuming that the transfer is between labor in the two countries, labor income in country I is enhanced and total income (and per capita income) available to the labor force in country II is further reduced. In fact, a study of remittances submitted by Greek emigrants indicated that the income, employment, and capital formation benefits to Greece from these remittances were substantial, while the costs of the emigration itself to Greece were limited (see Glytsos, 1993). More recently, the top four remittance-receiving countries in 2014 were India (estimated to have received $70.4 billion), China ($64.1 billion), the Phillipines ($28.4 billion), and Mexico ($24.9 billion). In 2014, an estimate in total was that developing countries received $436 billion in remittances. Developed countries also, of course, receive immigrant remittances ($146 billion in 2014).9 For comparison purposes, remittances received by the developing countries were almost 3 times the amount of official foreign aid received by these countries. A second issue is the nature of the immigration. We have assumed so far that the immigration is permanent, not temporary. A temporary worker, such as a Polish asparagus worker in Germany in the earlier vignette, is often called a guest worker. In the preceding analysis, all workers were assumed to be identical and the new immigrant thus received the same wage-benefit package as the domestic worker. This is not an unrealistic assumption because many countries do not permit employers to discriminate against permanent page 249immigrants. A two-tier wage structure is thus not possible. However, these restrictions do not often hold for guest workers or seasonal migrants.

Immigrant Remittances

A neglected economic feature in the immigration debate (both with respect to legal immigration and illegal immigration) is the flow of funds that occurs from the immigrants to their relatives back in their home countries. These flows can have significant effects on the countries from which the migrants originated. A 2011 set of estimates of the World Bank suggested the magnitude and impact of these flows. Immigrant remittances were estimated to be $416 billion during 2009, with $307 billion of that amount going to developing countries. However, these were only the recorded flows. In fact, unrecorded flows in 2009 to the developing countries were thought to be at least 50 percent larger than the recorded flows, which implies a total annual flow of about three-quarters of a trillion dollars [$307 billion + (1.50)($307 billion) = $768 billion]. In fact, even using only the recorded flows, the remittances were the second largest item in external funds received by developing countries (behind FDI). The funds were considerably larger than the amount of foreign aid received from developed countries. For specific countries as examples, data indicate that in 2008, Bangladesh received $9.0 billion in remittances and $2.1 billion in aid, Brazil received $5.1 billion in remittances and $500 million in aid, and the Dominican Republic received $3.6 billion in remittances and only $200 million in aid. It has also been estimated that remittances to Mexico were equivalent to 2.8 percent of Mexico’s GDP in 2008. Research suggests that remittance flows from the United States to Mexico are influenced by a number of factors including social capital, exchange rates, interest rate differentials, income, and proximity of migrants to Mexico. Interestingly, illegal immigrants to the United States from Mexico seemed more likely to send funds back to their families than did legal immigrants to the United States from Mexico. Remittances of this size can clearly benefit the recipient countries. An estimate by the World Bank is that such remittances have reduced the poverty rate by almost 11 percentage points in Uganda, 6 percentage points in Bangladesh, and 5 percentage points in Ghana. Such funds help the recipients purchase consumer goods, housing, education, and health care. The effect also seems to be countercyclical—when the fund-receiving countries go into recession, for example, the inflow of remittances seems to increase (in contrast to regular private capital flows, which would decrease in that instance). In addition, when substantial labor migrates abroad, this out-migration can relieve some of a labor surplus in the sending country and put upward pressure on wage rates. The sizeable level of remittances does not necessarily imply that the migrant outflow from the home countries is therefore a positive force for those countries, however. When the migrants leave, they often take substantial human capital with them because the migrants can be high-skilled workers. The tax base in the labor-sending countries is also being eroded when the workers leave—one estimate was that in 2001, immigrant Indians in the United States were equivalent to 0.1 percent of India’s population but equivalent to 10 percent of the national income of India. This fact meant that India’s lost tax revenue was perhaps equal to 0.5 percent of its GDP. In addition, large remittances into a country can lead to a rise in the value of that country’s currency and thus to a reduction in the country’s ability to export. Further, the inflow of funds may have an adverse impact on the work effort of the family members receiving the funds and thus reduce economic growth. In summary, the size of immigrant remittances presently being transmitted is substantial. There are positive and negative effects associated with the migration flow and with the remittances, and the net impacts on the home countries receiving the funds will vary from case to case. In any event, in today’s world, these flows and their impacts clearly need to be included in any analysis of labor migration. Sources: Dilip Ratha, “Remittances: A Lifeline for Development,” Finance and Development 42, no. 4 (December 2005), pp. 42–43; “Sending Money Home: Trends in Migrant Remittances,” Finance and Development 42, no. 4 (December 2005), pp. 44–45; Gordon H. Hanson, “Illegal Migration from Mexico to the United States,” Journal of Economic Literature 44, no. 4 (December 2006), p. 872; Kasey Q. Maggard, “The Role of Social Capital in the Remittance Decisions of Mexican Migrants from 1969 to 2000,” Federal Reserve Bank of Atlanta Working Paper 2004–29, November 2004; The World Bank, Migration and Remittances Factbook 2011, 2nd ed., obtained from www.worldbank.org. If migrant labor is not perceived as homogeneous with domestic labor, it is possible for the owners of capital in the recipient country to gain without reducing the income of domestic labor (see Figure 5). If employers can discriminate against the migrant worker, they will hire L1L2 short-term guest workers at the new market-clearing wage, 0W2, subsidize the page 250initial level of domestic workers by the amount of the total wage difference, W2W1AB, and gain area ABC. In this instance, country II clearly benefits because the permanent domestic labor force is no worse off and the owners of capital are clearly better off. It is not surprising that there is less opposition to temporary immigration than permanent migration, and there seemed to be none in the earlier asparagus example. It also is not surprising to see home labor discourage even seasonal labor immigration if it perceives that short-term migration keeps average wage rates fixed in the presence of rising production and product prices.

We need to make some final observations about the nature of the migrant and the implications of migrant characteristics on both countries. The assumption that workers are homogeneous is certainly not true in the real world, and the welfare implications that accompany migration can vary as a result. The labor force in each country possesses an array of labor skills ranging from the untrained or unskilled to the highly trained or skilled. For this discussion, let us assume that each country has only two types of labor, skilled and unskilled. The implications of out-migration on the home country vary according to the level of skill of the migrants. The traditional migrant responding to economic forces tends to be the low-skilled worker who is unemployed or underemployed in the home country and who seeks employment in the labor-scarce country with the higher wage. The motive for the migration is not only the higher wage in the host country but also the greater probability of obtaining full-time work, along with other considerations. The movement of low-skilled workers based on expected income differentials has effects on the two countries that are consistent with our previous analysis. Total world output rises, output falls, and average low-skilled labor income rises both absolutely and relatively in the home country, and output rises and average income of low-skilled labor falls both absolutely and relatively in the host country. It is important to note that the return to skilled labor in the host country, like capital, is likely to rise. page 251 The host country may also experience increased social costs through larger expenditures for human safety-net programs (unemployment transfers, education, housing and health subsidies, etc.) as the number of unskilled workers increases relatively and absolutely. Because the unskilled worker tends to suffer greater employment instability, an increase in the relative number of unskilled workers is generally linked to higher social maintenance costs. An increase in these indirect costs results in higher taxes, therefore reducing the net gain for owners of other factors such as capital. The reduction in average low-skilled wages, including the concomitant increased taxes, is thus greater than suggested by the fall in the market wage alone. It is not surprising that most countries attempt to control the immigration of low-skilled workers. In an attempt to avoid some of the indirect social costs of this immigration, several European countries such as Switzerland have in the past adopted guest worker policies that allow low-skilled labor to immigrate for short periods of time, but the workers do not qualify for citizenship and can be required to leave the country at the government’s request. The movement of skilled labor, especially between developing and industrialized countries, is a relatively recent phenomenon. However, an increasing number of highly educated people [economists(?), physicians, research scientists, university professors, and other skilled professionals] are leaving the developing countries for the United States, Canada, and western Europe—a movement often referred to as the brain drain. Higher salaries, lower taxes, greater professional and personal freedom, better laboratory conditions, and access to newer technologies, professional colleagues, and the material goods and services found in these countries explain this movement of labor. In many cases, the person had received formal training in the industrialized country and found it difficult to readjust, at least professionally, to life in the home country. From an economic standpoint, if markets are working and labor is paid its marginal product in both countries, the analysis of skilled-labor movements is similar to that of unskilled labor, except for the differences in magnitude connected to the difference in marginal products. It is possible, however, that skilled labor is in such short supply in the home country that the loss of these workers leads to a fall in per capita income, not an increase. The opportunity cost to the home country may be even larger than indicated by the market wage if the skilled worker generates other positive benefits (externalities) for the home country such as a general improvement in the level of technology. In addition, to the extent that the home country has subsidized the education of these people (i.e., invested in their accumulation of human capital) the out-migration represents a loss of scarce capital on which a reasonable social rate of return was expected. Finally, the cost to the home country is even greater if markets are distorted by government regulation in a way that the individual was receiving something less than the free-market wage. In that event, the wage formerly received by the worker understated the true market value of the worker. The opposite is true in the recipient country. The productivity of the immigrant skilled worker is relatively higher, the possibility of positive externalities is greater, and expected indirect social costs are lower than with the low-skilled migrant. In addition, the inflow of the skilled professional reduces the domestic price of nontraded services such as medical care. In this case, the pressure against immigration will come from professional labor groups, not from the overall labor force. In general, however, most industrialized countries have done little to restrict the immigration of skilled workers, and in some cases have made it easier for skilled workers to obtain work visas than is the case for unskilled workers. The developing countries are in a quandary. The migration of skilled labor often represents a substantial static and dynamic cost to them. Because the combination of externalities, market wage distortions, and the opportunity cost of the human capital investment frequently exceeds the income paid to the skilled worker, countries are often inclined to restrict the out-migration of skilled labor. Until recently, for example, restrictions of this kind were common in eastern Europe. However, the loss in personal freedom associated page 252with labor movement restrictions makes such restrictions unappealing. Restriction of personal freedoms may also lead to lower productivity and a loss of professional leadership and entrepreneurship, which is important to these countries as they undergo economic reforms. Several policies can be directed toward removing market imperfections: (1) paying skilled labor its marginal product, (2) subsidizing professionals so that their income reflects their true social value including externalities, (3) taxing out-migrants or requiring remittances from them to cover at least part of the investment in human capital, (4) guaranteeing employment and high-quality jobs to those who return home following training abroad, and (5) appealing to the nationalism of the skilled worker. These policies may be more attractive than the restriction of free movement between countries. While the movement of skilled labor from developing countries to the industrialized countries may lead to an increase in efficiency and world output in the static sense, it contributes to increased divergence of income between low-income and high-income countries. In addition, the loss of this very scarce resource alters the dynamics of change in the developing countries. Thus, the correct policy response is not clear. The answer to the question, Which is larger?—the social cost reflected in the loss in personal freedoms caused by emigration restrictions or the social cost associated with free outward movement of labor—must be sought beyond economic paradigms. In the end, individual freedom of movement may well dominate any economic considerations.

Immigration and the United States—Recent Perspectives

We cannot leave this analysis of international labor movements without a brief discussion of the large volume of research related to the economic impact of immigration on host countries in general and the United States in particular.10 Inasmuch as this research is directed toward an examination of immigrant performance, impact on host-country labor markets, and the likely impact of immigration policy, a brief presentation of some key findings is a fitting way to conclude our discussion of the economic implications of international labor movements. What emerges very clearly in the case of the United States is that the economic characteristics of immigration have been changing in recent years both with respect to initial migrant earning performance and the broader, longer-term implications for the economy in general. Up through the 1970s, based on the stylized facts regarding immigration in the first half of the century, it was widely accepted that although immigrants as a group were initially in an economically disadvantaged position, their earnings soon caught up with the earnings of those domestic workers with similar socioeconomic backgrounds and eventually surpassed them. What was interesting was that this adjustment took place in a relatively short time, within 10 to 20 years on average, and appeared to have little or no adverse impact on the domestic labor market. Much of this shift can be traced to the fact that U.S. immigration laws were changed in 1965 toward favoring immigrants with existing family ties to residents of the United States and away from a focus on the skill levels of the immigrants. Only about 15 percent of new green cards recently issued were awarded for work reasons, rather than for family relationships, humanitarian causes, and other reasons.11 Research by George Borjas (1992; 1994, p. 1686) also indicated that the origin of U.S. immigrants had changed, with a marked increase in the proportion coming from developing countries. Concomitant with this change in country of origin, there was a decline in the immigrants’ skill levels over much of the postwar period. Borjas therefore concluded that it is not likely that the more recent wave of immigrants will continue to obtain wage parity with domestic workers of similar socioeconomic backgrounds.12 This suggests not page 253only that they will likely have a heavier participation rate in U.S. welfare programs but also that this differential will carry over into second-generation wage and skill differences, which will be reflected in widening ethnic income differences within the overall labor market.13 In fact, some research suggests that immigrants with less than a high school education are a net cost to the United States, in the sense that the value of the public services they use exceeds the taxes they pay. For high-skilled immigrants, the reverse is true, in that the taxes paid exceed the cost of the services used.14 There is also weak evidence that the increasing numbers and declining skill levels of immigrants may have contributed to the relative decline of domestic unskilled wages in the 1980s. For example, Borjas, Richard Freeman, and Lawrence Katz (1992) concluded that perhaps one-third of the 10 percent decline in the relative wage of high school dropouts from 1980 to 1988 could be explained by immigration flows. If these trends are indeed the case and continue into the 21st century, there will likely be far-reaching and long-lasting effects on the labor force, net welfare costs, and income distribution in the United States. Countries that are able to effectively control the skill characteristics of the new migrants will be able to negate some of the aforementioned negative effects. However, in the United States, for example, the H–1B visas that are awarded to skilled potential immigrants with a bachelor’s degree or higher are strictly limited by an annual quota system and are applied for far in excess of the number to be awarded.15 Such a limitation can potentially harm domestic workers in that, for instance, a recent study suggested that U.S. cities with the largest inflow of highly skilled workers had the greatest increases in wages for American-born college graduates in those cities.16 However, this interesting result raises an issue which warrants further investigation. It is not surprising that immigration policy is a “hot topic” in government circles in Washington, DC. Adding to the discussion is the emerging view that, without continued immigration, the United States may soon see a marked slowdown in the growth of its labor force as its population gets older. IN THE REAL WORLD: IMMIGRATION AND TRADE Some recent literature has been concerned with the links that may exist between the stock of immigrants in a country and the trading and other relationships of the host country with the home countries of the immigrants. This literature makes the broader point that labor movements between countries affect not only labor markets per se in the receiving and sending countries of the labor but also have secondary impacts on a range of other economic variables. An example of work that links immigration to trade is provided in a paper by Roger White (2007). In this paper White employed a gravity model (see the earlier discussion in Chapter 10, pages 195–96) in an attempt to explain various influences on U.S. trade. He empirically investigated the trade of the United States with 73 trading partners for the time period 1980–2001. Gravity model equations were run with the dependent variables alternately being U.S. total trade, U.S. exports, and U.S. imports. Standard independent variables for the gravity model such as the GDP of the United States and the GDPs of partner countries were included, as were exchange rates and distance. Two of the other independent page 254variables were (1) whether or not there existed a free-trade agreement between the United States and any given partner (which would, other things being equal, increase the amount of trade) and (2) whether or not English was an official language of the partner (which would also increase trade). The independent variable of prime interest was the number of immigrants from the given partner country who were living in the United States. The central hypothesis in this immigrant-trade literature is that trade will be enhanced between the sending country of the migrants and the host country because of, for example, the desire of the immigrants to consume products to which they are specifically accustomed and that might not be produced with identical characteristics in the host country. In addition, social and business contacts and networks between the immigrants and residents/firms in the home country may make it easier and less costly to continue operating within those established relationships than to develop a whole new set of relationships (that is, transaction costs may be kept lower than otherwise would be the case). The regressions that were run by White generally produced no surprises for the traditional variables. Of importance for this chapter was the finding that trade volume was indeed increased by the presence of immigrants. For the full sample of 73 countries, White estimated that, other things being equal and on average, a 10 percent increase in the stock of immigrants in the United States from any given trading-partner country would increase U.S. imports from that country by 1.3 percent and would increase U.S. exports to that country by 1.1 percent. Further, a new finding—one that had not been uncovered in previous studies—emerged when White disaggregated the sample into high-income, medium-income, and low-income partner countries. For the low-income partners, a 10 percent increase in the stock of immigrants from any given country would increase U.S. imports from that country by 4.66 percent and would increase exports to that country by 1.47 percent. Stated in more concrete terms, for example, he estimated that the average U.S. immigrant from China adds $11,442 annually to the U.S.–China total trade, while examples of corresponding numbers for other countries’ immigrants to the United States are $10,724 for Bangladesh, $6,252 for Nigeria, $718 for Nicaragua, and $164 for Vietnam. However, and importantly, there did not appear to be any trade-increasing effects of increased immigration from high-income and medium-income trading partners. Thus, the overall impact of the stock of immigrants on trade was in effect accounted for by immigrants from low-income countries and not by immigrants from the medium- and high-income countries. These are intriguing findings that clearly call for more investigation as to the reasons for their occurrence. An extension of this work into the broader area of culture and values has been explored by White and Bedassa Tadesse (2008). They investigated what they labeled as the “cultural distance” between countries and the effects of that distance on trade flows, and they employed data from World Values Surveys and European Values Surveys to do so. These surveys involve the completion of questionnaires by representative samples of the population in many countries.* In the White/Tadesse paper, the survey results used were from the 1998–2001 time period, and they contained information on politics, religion, gender roles, ethical considerations, and other such matters. White and Tadesse constructed two indexes for the United States and for each of 54 trading partners, and, for each index, a greater difference in the given index between any two countries indicated greater “cultural distance.” Using these indexes and 1997–2004 trade data and other relevant economic information, White and Tadesse then ran gravity model regressions with U.S. exports and U.S. imports used alternately as the dependent variable. Normal results were generally obtained for the signs of traditional independent variables, such as GDP and the existence of a trade agreement. The independent variable of the stock of immigrants in the United States from any given country yielded statistically significant positive signs regarding trade, as in the White study discussed in the previous paragraphs. With respect to the cultural indexes, both the export and the import regression yielded statistically significant negative signs for one of the two indexes, meaning that a greater cultural difference between the United States and any given trading partner resulted in, other things being equal, less trade between the United States and that partner. However, for the other cultural index, the expected negative sign occurred for U.S. imports but not for U.S. exports to the given partner (in fact, that latter result was a positive sign). Hence, although cultural distance does seem to play a role in some way with regard to the volume of trade, further empirical (as well as theoretical) investigation appears to be necessary. *Further information on these surveys is available at www.worldvaluessurvey.org. Sources: Roger White, “Immigrant-Trade Links, Transplanted Home Bias and Network Effects,” Applied Economics 39, no. 7 (April 20, 2007), pp. 839–52; Roger White and Bedassa Tadesse, “Cultural Distance and the U.S. Immigrant-Trade Link,” The World Economy 31, no. 8 (August 2008), pp. 1078–96. page 255

Immigration into the United States and the Brain Drain from Developing Countries

Several studies have shed light on the type of labor that decides to emigrate to the United States and the impact of immigrants on the U.S. economy. While there is considerable debate regarding the Borjas claim that current U.S. immigrants are relatively less skilled than their earlier counterparts (and thus that current migrants are less likely to have a positive impact on the economy than their predecessors),* it appears clear that the typical person who has emigrated from most developing countries in the past is relatively skilled. In 1999 William J. Carrington and Enrica Detragiache presented the results, using 1990 census data, of an examination of the educational background of the stock of developing-country emigrants (not the flow of migrants, which Borjas was examining) over 25 years of age who now reside in the United States.† The first striking result in the study was that individuals with no more than a primary education (zero to eight years of schooling) accounted for only about 7 percent of the total immigrants (i.e., about 500,000 of the total of 7 million immigrants). Approximately 53 percent (3.7 million of the 7 million) were persons from other North American countries (which included Central American and Caribbean countries in the Carrington and Detragiache definition) who had at most a secondary education. Most of these individuals were from Mexico. Almost 1.5 million immigrants (21 percent) were highly educated individuals with a tertiary level of schooling (more than 12 years) from Asia and Pacific countries. (Note: this “highly educated” measure does not include international students in the United States, who were excluded from the “immigrant” definition.) In addition, although small in number (128,000), 75 percent of immigrants into the United States from Africa consisted of highly educated individuals. More than 60 percent of migrants from Egypt, Ghana, and South Africa had a tertiary education, as did 75 percent of migrants to the United States from India. Immigrants from China and South American countries were about equally divided between the secondary and tertiary education levels. Mexico and Central American countries thus appeared to be an exception in that most of the migrants from those countries had education only through the secondary level. An important point to make is that, in general, individuals who emigrate to the United States tend to be better educated than the average person in their home countries. Further, the migrants often represent a sizeable portion of the similarly skilled workforce in their own countries. Carrington and Detragiache present some truly startling statistics in this regard. They calculated the stock of immigrants of a given education level in the United States from any given country and then divided that number by the size of the population of the same education level who remained in the home country. For example, at the tertiary-education level, the number of Jamaican immigrants in the United States divided by the size of the Jamaican population with tertiary education gave a figure of 70 percent. While the number of Jamaican immigrants is relatively small in absolute terms and the percentage of the Jamaican population with tertiary education is likewise small, this figure gives concrete force to the notion of brain drain from developing countries. Other (small) developing countries also had high numbers with regard to the tertiary-education level—Guyana (70 to 80 percent), The Gambia (60 percent), and Trinidad and Tobago (50 to 60 percent). El Salvador, Fiji, and Sierra Leone had ratios greater than 20 percent. For many countries in Latin America, the ratios that were the highest were those with respect to secondary education rather than tertiary education [e.g., Mexico (20 percent), Nicaragua (30 percent)], but, even so, their magnitude indicates a substantial outflow of skill. This loss of tertiary-level (and secondary-level) individuals cannot help but impede the economic and social progress of source countries spread throughout the world. However, recent research suggests some mitigating factors. For example, brain drain scientific personnel appear to interact with peers in their home countries, sharing ideas and increasing the flow of innovation from developed to developing countries. *See George Borjas, Heaven’s Door (Princeton, NJ: Princeton University Press, 1999); Jagdish Bhagwati, “Bookshelf: A Close Look at the Newest Newcomers,” The Wall Street Journal, September 28, 1999, p. A24; Spencer Abraham, “Immigrants Bring Prosperity,” The Wall Street Journal, November 11, 1997, p. A18; “Immigrants to U.S. May Add $10 Billion Annually to Economy,” The Wall Street Journal, May 19, 1997, p. A5; “The Longest Journey: A Survey of Migration,” The Economist, November 2, 2002, p. 13 (where an estimate is presented that first-generation migrants to the United States impose an average net fiscal loss of $3,000 per person while the second generation yields an $80,000 net fiscal gain per person); “Give Me Your Scientists,” The Economist, March 7, 2009, p. 84. †William J. Carrington and Enrica Detragiache, “How Extensive Is the Brain Drain?” Finance and Development 36, no. 2 (June 1999), pp. 46–49.

Summary

This chapter discussed various aspects of international factor movements between countries. Causes and consequences of international mobility of capital and of labor have been examined, and particular attention has been devoted to some implications for international trade and relative factor prices. Movements of factors of production have received relatively little attention in the literature on international economics compared with movements of goods and services, and a systematic and comprehensive framework incorporating the many facets of these movements remains to be formulated. In addition, judgments on the welfare and development implications of factor flows differ according to who is making the assessment and to the weights placed on various objectives. As capital and labor mobility become more prominent in the world economy in the future, it will increasingly become necessary to investigate further the causes, the consequences, and the policy implications of the international movements of factors of production.

International Economics (Chapters 6) - Krugman, P R; Obsterfeld, Melitz, M J (2021, Twelfth Edt.)

Chapter 6:

Page updated

Google Sites

Report abuse