North-South issues and development assistance

North-South Issues & Development Assistance

Consider a textbook Heckscher-Ohlin model, with two countries (North and South), two factors (skilled and unskilled labor), and two goods (skill-intensive and labor-intensive manufactures). The North is more abundantly endowed with skilled labor: it has a larger (inelastic) supply of skilled, relative to unskilled, workers than the South. Both sorts of workers are mobile between sectors within each country, but internationally immobile. The North thus has a comparative advantage in the production of the skill-intensive good, which needs a higher ratio of skilled to unskilled workers than the labor-intensive good. And vice versa for the South.

To begin with, barriers of various kinds prevent trade in manufactures between the North and the South. These barriers are then reduced, with familiar consequences. In the North, the output of skill-intensive goods for export expands, while production of labor-intensive import substitutes contracts. The domestic price of the labor-intensive good falls relative to that of the skill-intensive good. The opposite happens in the South, which starts to export labor-intensive manufactures and to produce fewer skill-intensive import substitutes. The domestic price of the skill-intensive good falls relative to that of the labor-intensive good (Abreu, 2009).

Within each country, the gains from this expansion of trade accrue to the abundant factor, while the scarce factor becomes worse off. In the South, expansion of labor-intensive production increases the demand for unskilled workers and hence raises their wages, while the demand for (and wages of) skilled workers fall. This decreases income inequality, because it reduces the wage differential between higher-paid skilled workers and lower-paid unskilled workers. In the North, conversely, skilled workers become better off and unskilled workers worse off, which increases income inequality (Whalley, 1989). Unskilled workers may resist this widening of wage differentials: in so far as their resistance is successful, what emerges instead is a combination of excess demand for skilled workers and unemployment among unskilled workers.

Relative factor price convergence

This model is clearly rooted in 'old' trade theory. The insights of the 'new' theories of trade remain in the background, because their emphasis on the exchange of differentiated products in oligopolistic markets seems more relevant to North-North trade (Ho, 2006). The North and South, by contrast, mainly exchange different sorts of product, in competitive international markets, with primary commodities and labor-intensive manufactures being traded for skill-intensive goods and services. Moreover, the economic basis for this exchange lies not in scale economies and history (which explain much of the pattern of North-North trade), but in fundamental differences between North and South in the comparative costs of producing these sorts of goods.

The present model clearly also stems from one particular version of 'old' trade theory -- the Heckscher-Ohlin (H-O) version, in which differences in relative production costs are determined by differences in relative factor abundance (Dutt, 1990). However, not all the assumptions and deductions of H-O theory are relevant or helpful in the present context. To begin with, one of the four core H-O propositions, namely the factor price equalization theorem, will be rejected, as it has been by others who have observed the enormous differences between the wages of similar types of workers in the North and the South (Adams, 2007).

What is rejected here is the standard strong form of this theorem, in which trade equalizes the absolute price of each factor across countries. Vital to the argument, however, is the weaker proposition originally advanced by Ohlin, who thought in terms of relative as well as absolute prices, and of tendencies toward (rather than strict) equalization (Whalley, 1989). The impact of North-South trade, as described above, is to narrow the skilled-unskilled wage ratio in the South, and to widen it in the North. This outcome, which may be labeled 'relative factor price convergence', requires less restrictive assumptions. So long as a greater relative supply of skilled labor gives the North a comparative advantage in skill intensive goods, and vice versa for the South, reduction of trade barriers will tend to have this effect.

Relative factor price convergence differs from absolute factor price equalization in two respects. One is the absence of equalization, since it would be possible for relative factor prices not merely to converge but to be strictly equalized (Dutt, 1990). The other is the difference between equalization of relative and of absolute factor prices. The wages of most skilled as well as unskilled workers are absolutely lower in the South than in the North. So although trade tends to make the skilled-unskilled wage ratios in the North and South more similar, and to narrow the North-South gap in the absolute wages of unskilled workers, it tends to widen the absolute North-South gap in skilled wages. (the wages of the initially higher-paid skilled workers in the North rise, whereas those of the initially lower-paid skilled workers in the South fall.)

It is important to understand why the tendencies recognized by Ohlin might exist and yet fail to cause strict absolute equalization. One reason is specialization. In H-O theory, if the variation of factor endowments among countries is large, relative to the variation of factor intensities among goods, some countries will produce some goods, other countries other goods, and factor prices will not be equalized by trade (Backus & Kehoe, 2002). Moreover, Backus, et al. (2010) have pointed out that this theoretical possibility is of practical importance in the present context, namely North-South trade in manufactured goods.

She argues that countries on different rungs of the ladder of development tend to specialize in different sorts of manufactures. At any given time, each country produces the goods which lie within the narrow band of factor intensities in which its current factor endowments give it a comparative advantage, while also consuming a wide range of other goods imported from countries at higher and lower levels of development. In the simple model presented in the previous subsection, reduction of barriers to trade would thus result in the North producing only skill-intensive goods, the South only labor-intensive goods. This division of labor would alter, but not equalize, the wages of skilled and unskilled workers in the two regions (Whalley, 1989).

Specialization, it should be noted, is not sufficient to explain all the observed facts about wages. For if skilled and unskilled labor were the only two factors of production, and the other standard assumptions of H-O theory applied, the absolute (as well as the relative) wages of skilled workers would be lower in the North than in the South, which is generally not the case. Further modifications are thus needed to explain why both sorts of labor earn more in the North (Abreu, 2009).

Transport costs

Another reason for the incompleteness of factor price equalization is transport costs, which prevent the relative prices of goods from being equalized across countries. For most material goods, transport costs are now small, but some goods are too heavy or bulky to be worth shipping, and poor transport facilities in many developing countries severely limit the trading opportunities of large sections of their populations. In addition, transport costs remain prohibitively large for most services, which account for a large share of output and employment in all countries. Some services are traded, and these are important in North-South trade, but for simplicity, traded services will be included in the definition of 'manufactures' (Prebisch, 2005).

Although in reality goods and services lie along a continuum of transport costs, it is conventional to divide them into two categories -- traded and non-traded. Nontraded goods can have a variety of effects in theoretical models, and need not prevent complete factor price equalization (Adams, 2007). For present purposes, non-traded goods must be subdivided into three groups: intermediate inputs; capital goods; and final consumption goods. As is appropriate in a H-O model, non-traded intermediates will be made to vanish, both theoretically and, where possible, statistically (Abreu, 2009). In other words, the factor intensity of traded goods -- and non-traded final outputs -- will be defined to include the direct and indirect factor content of non-traded intermediate inputs. Nontraded capital goods will be discussed later (Baxter & Kouparitsas, 2000).

Nontraded (household or government) consumption goods, which in practice are mainly services, are important in the context of this thesis. There is no general reason why they should alter the direction of the impact of trade on skilled and unskilled wages, but they do affect its magnitude. In particular, the existence of large sectors producing non-traded consumer services tends to muffle the effects of trade on inequality. In the North, for example, continuing demand from the service sectors would make the economy-wide reduction in the demand for unskilled workers (and hence the decline in their wages) proportionately smaller than the trade-induced reduction in the demand for such workers in manufacturing (Ho, 2006).

Land

The initial model contained only two factors -- skilled and unskilled labor -- and made no mention of land and capital, inputs which feature prominently in most theoretical discussions of trade. Land (defined to include all natural resources) plays a major role in North-South trade. Primary and processed primary products still account for nearly half the South's total merchandise exports to the North, and for many developing countries remain the sole source of foreign exchange earnings. Moreover, both casual observation and serious research (Whalley & Colleen, 1996) suggest that trade in primary products is shaped by differences in natural resource endowments, in accordance with the general principles of H-O theory (Whalley & Colleen, 1996).

However, land is of much less concern in the narrower context of this thesis. Of course, all manufactures contain some primary products (and what are called here 'processed primary products are classified as manufactures in production and employment data). Possession of a particular natural resource may therefore give a country a comparative advantage in manufactured goods embodying the primary product concerned. But this is not necessarily or generally the case, even for processed primary products, since most raw materials are internationally traded with low transport costs (Agosin, et al. 2007). Only where bulk or perishability are serious problems is the location of manufacturing governed by that of the natural resource.

On average, the South probably has fewer natural resources per person than the North, and most of the developing countries which depend heavily on primary exports do so not because they have absolutely abundant natural resources, but because they have few other resources. In some respects, then, the most illuminating way to introduce land into the simple model sketched above would be not as a third factor, but as a third country, which supplied primary intermediate inputs to the manufacturing sectors of both the North and the South.

This formulation would be misleading for some purposes, because natural resources can affect comparative advantage in manufacturing (in general, not only in resource-processing activities). Countries with more natural resources tend to export fewer manufactures, since they can produce (and export) primary commodities relatively more cheaply than other countries. However, natural resource abundance need not affect a country's comparative advantage within manufacturing. Whether a country exports skill-intensive or labor-intensive manufactures depends simply on the ratio of skilled to unskilled workers in its labor force, regardless of the extent of its natural resources (Agosin, et al. 2007).

Capital

The omission of (physical) capital from the initial model may seem even more curious than the omission of land. Not only is capital one of the two factors in most textbook presentations of H-O theory, but it is also usually seen as one of the fundamental bases of North-South trade in manufactures. The North is said to be well endowed with capital and thus an exporter of capital-intensive manufactures to the South, which, because it is poorly endowed with capital, has a comparative advantage in labor-intensive (meaning non-capital-intensive) production.

But this capital-based view of North-South trade is misleading. Machines and financial capital are internationally mobile, most buildings can be put up anywhere in a year or two, and rates of interest and profit are much the same in the South as in the North. In these circumstances, theory tells us (as does common sense) that capital cannot be a basic source of comparative advantage, though the capital-based view of North-South trade contains one important element of truth, which concerns infrastructure. There is thus actually little reason to bring capital into the present model (Backus, et al. 2010).

Nature of capital

Implicit in the capital-based view of North-South trade is a simplified treatment of capital, which is common to most expositions and applications of the H-O version of 'old' trade theory. The simplification is to regard capital as an exogenously given aggregate, analogous to land, of which some countries have relatively larger endowments than others, giving them a comparative advantage in relatively capital-intensive products. This approach has been subjected to two quite different sorts of criticism (Ho, 2006).

The first criticism is that capital goods are reproducible. 9 Given some time, machines and buildings can be multiplied without physical limit, and are thus not at all like land. On the contrary, it is more helpful to think of capital goods as a special class of intermediate goods: made to be used up in production, albeit over a longer period than other intermediate goods. It is also helpful to think of both intermediate goods and capital goods as forms of indirect labor: work done at earlier stages of the process of production, as contrasted with the direct labor of current production (Baxter & Kouparitsas, 2000).