North-South Trade and Basic Needs

(1)

NOT FOR QUOTATION WITHOUT PERMISSION

OF THE AUTHOR

NORTH-SOUTH TRADE

AND

BASIC

NEXDS

Graciela Chichilnisky

February 1984 CP-84-04

Collaborative P a p e r s report work which has not been performed solely a t the International Institute for Applied Systems Analysis and which has received only limited review. Views or opinions expressed herein do not necessarily represent those of the Institute, i t s National Member Organizations, or other organizations supporting t h e work.

INTERNATIONAL INSTITUTE FOR APPLIED SYSTEMS ANALYSIS 2361 Laxenburg, Austria

(2)

(3)

FOREWORD

This Collaborative Paper is one of a series embodying the outcome of a workshop a n d conference on Economic Sruct-u~al Change: Analytical Issues, held a t IIASA in July and August of 1983. The conference and workshop formed part of t h e continuing IIASA program on Patterns of Economic Structural Change and Industrial Adjustment.

Structural change was interpreted very broadly: the topics covered included the nature and causes of changes in different sectors of t h e world economy, the relationship between international markets and national economies, and issues of organization and incentives in large economic systems.

There is a general consensus that important economic structural changes a r e occurring in t h e world economy. There are, however, several alternative approaches t o measuring these changes, t o modeling the process, and to devising appropriate responses in terms of policy measures and institu- tional redesign. Other interesting questions concern the role of the international economic system in transmitting such changes, and the merits of alternative modes of economic organization in responding to structural change.

All of these issues were addressed by participants in t h e workshop and conference. and will be the focus of the continuation of the research program's work.

Geoffrey Heal Anatoli Smyshly aev Ern6 Zalai

(4)

(5)

This paper examines t h e role of the international m a r k e t in mediating North-South relations and analyzes how t h e market works in distributing t h e gains from trade. It is argued t h a t t h e international market does not always provide an adequate engine of growth for t h e South if t h a t region specializes in labor-intensive products. The South's export sector m u s t be carefully balanced with o t h e r domestic sectors t o avoid harming t h e economy a s a whole.

Any excessive expansion of labor-intensive exports or raw materials, even if accompanied by a n expansion in international d e m a n d m a y affect domestic markets and t h e distribution of income in t h e South in ways t h a t conflict with sustainable development. especially when this is measured in t e r m s of t h e satisfaction of basic needs for t h e majority of t h e population. The conditions under which this may o c c u r are quite general. They a r e consistent with perfect market behavior b u t require t h a t important features of t h e North-South relationship, including differential characteristics of technologies a n d factor markets in t h e two regions. be introduced into t h e analysis. The paper suggests alternatives to export-led policies, which balance domestic a n d international sectors of t h e South's economy and a r e conducive t o sustained d e v e l o p m e n t and t h e satisfaction of basic needs. An appendix provides a model of North-South trade t h a t h a s been econometrically t e s t e d for t h e trade between Sri Lanka and t h e UK. The appendix also includes a computer program for simulating t h e model and sarnple computer runs t h a t reproduce, in practical terms, t h e model trade policies discussed in the paper.

(6)

(7)

NORTH-SOUTH

TRADE AND

BASIC

NEEDS*

Graciela Chichilnisky**

1. BASIC

NEEDS AND

INTERNATIONAL MARKEX3

The concept of basic needs has received widespread attention i n t h e development literature over t h e last t e n years, since its introduction in t h e early seventies in t h e Bariloche Model. This model was produced in t h e Funda- cion Bariloche, Rio Negro, Argentina over t h e period 1972-76, under t h e auspices of t h e International Development Research Center of Canada; see e.g.

Herrera e t al. (1977) and Chichilnisky (1977b). The model was first presented a t IIASA in 1975, a t a conference chaired by Tjalling Koopmans. The model itself, and its introduction of t h e satisfaction of basic needs as a goal of development has been subsequently discussed by several authors; see e.g.

Nordhaus (1975), Streeten (1978). and Hopkins e t al. (1976). h 1976 t h e basic needs concept was brought t o t h e attention of the wider international develop ment community during t h e World Employment Conference of t h e Interna- tional Labor Organization

(ILO).

_Anumber of papers and books studying basic needs development strategies followed: see e.g. Richards and Leonor (1982).

These refined and extended t h e ideas introduced in t h e Bariloche Model, which defined a basic needs development strategy as one aiming for minimum levels of per capita consumption of food, shelter, education, and health. The Bari- loche Model explored whether t h e goal of satisfying basic needs was within t h e reach of t h e developing countries, within their existing constraints on resources and population. The answer was positive, but qualified: many reforms appeared to be needed for regions such as Africa to reach this goal.

The model simulated these reforms, planning development paths t o reach these goals under different scenarios and for different regions of t h e world.

*This research w a ~ supported by the International Institute for Applied System Analysis, h e n - b u g , Austria, and by the US National Science Foundation under Grant No. SES 7814050.

*Folumhia University, New York a n d United Nations Institute for Trai- and Research. New Ymk.

(8)

However, because of t h e model's large scale, these economic plans were highly aggregated and left large areas of economic behavior unexplained. For instance, little attention was paid to domestic or international m a r k e t behavior.

Markets a r e powerful economic forces in developing and industrial coun- tries alike. In particular, the international m a r k e t has become an iricreas- ingly powerful force in the world economy, even affecting those countries with a certain degree of central planning, following the unprecedented growth of international t r a d e from 1945 t o the mid 1970s.' Large segments of t h e

GDP

of inddstrial and developing countries are now linked t o international markets.

Yet in spite of t h e increasing degree of internationalization of our economies, the concept of basic needs has been essentially viewed in m u c h of the lit&ature as a domestic issue. The Bariloche model discussed basic needs in t h e international context, but did not analyze in detail the impact'of international m a r k e t s on t h e satisfaction of basic needs, since it was fundamentally a planning model. The studies on basic needs strategies t h a t followed t h e Bariloche work have focused mostly on d o m e s t i c policies, and little has been said about t h e connection between basic needs and t h e internafional policies of developing countries. However, t h e international policies of an economy play a significant role in determining the level to which basic needs can be satisfied domestically.

This paber examines the missing link between international policies and basic needs policies, by discussing alternative trade theories and exploring the policies t h a t each suggests for encouraging the satisfaction of basic needs in a developing economy. The Appendix provides computer simulations of a model of North-South trade, and examines in practical terms the impacts of exports on domestic consumption, investment, employment, and wages. We find t h a t i n some cases t h e expansion of trade is favorable t o t h e satisfaction of basic needs b u t in other cases i t is not. We encounter results that may seem contrary t o existing notions of gains from trade and advantages derived from international specialization. We also question the effectiveness of t h e international market in transmitting growth from one region to another: we show t h a t an industrial expansion of the North may lead to a new market equilibrium with a larger volume of exports from the South, but may a t t h e same time reduce export revenues and t h e level of industrialization in t h e South.

Moreover, this increase in exports may be associated with a worsening of t h e distribution of- income within the South. Under such conditions, then,

"export-led" development strategies are not favorable to the satisfaction of basic needs nor t o wider development objectives. There is, therefore, a need for policy measures t h a t correct such conditions before gains from trade can be assured, and we discuss alternatives to export-led policies t h a t may be preferable until these conditions are corrected. These alternatives involve a more balanced view of development t h a t relies on domestic markets as well as the international market and concentrates on raising domestic productivity in crucial sectors of t h e economy.

The r e s t of this paper is organized as follows. Section 2 summarizes t h e standard neoclassical development and trade literature. which provides t h e intellectual support for much of current development policy in t h e North and South. a s well a s in t h e major international financial institutions. I t presents also a summary of criticisms of this body of thought, including some t h a t accept t h e basic paradigm. but seek t o cope with anomalies, and others t h a t reject the basic paradigm altogether. Section 3 outlines a North-South model which we developed and tested econometrically in the context of a UNITAR

(9)

project,2 and which is presented formally in the Appendix. This model remains within t h e framework of perfectly competitive markets, though i t introduces into this framework assumptions t h a t could be considered more realistic, and t h a t a r e in part suggested by some of the critical l i t e r a t u r e reviewed in t h e second section. Section 4 discusses t h e main results. We conclude with some policy inferences and recommendations. An Appendix summarizes t h e theoretical s t r u c t u r e of t h e model and discusses several empirical illustra- tions.

2.

A

BRIEF SURVEY OF TRADE THEORY

The impact of international markets on the functioning of domestic economies is now genarally acknowledged. However, recognition of this impact within t h e industrial economies is rather recent; i t largely emerged due t o the.changes t h a t took place during the seventies in the international markets for energy. By contrast, within the South t h e international m a r k e t h a s always been considered of fundamental importance to t h e domestic economies, because of the traditional reliance of many of these countries on exports of raw materials and labor-intensive products.

Over t h e last thirty years t h e r e has been developed in the North a consistent body of .formalized theory on the effect of trade on domestic economies. This theory b a s two main pillars: the results on gains from trade and international specialization, and t h e results on factor price equalization.

Both these results have been generalized, expanded, and applied very widely, and their conclusions have reached t h e noneconomist and become common and mostly undisputed knowledge within the h o r t h . These results a r e often used t o substantiate claims t h a t more trade is better for a& especially when t h e relative advantages in t h e trading regions a r e exercised. For t h e limita- tions of s u c h claims see Samuelson (1962).

Standard neoclassical trade theory is usually presented in the framework of the Heckscher-Ohlin model of two regions trading with each other. Each region produces two commodities using two factors of production, labor and capital. One good is produced in a more labor-intensive way t h a n t h e other.

These two countries a r e identical in all respects except for a difference i n factor endowments t h a t leads to differentials in product a n d factor prices i n each region.

In

isolation, t h a t is prior t o trade, the labor-intensive product i n t h e labor-abundant country h a s a lower price than the s a m e product in the o t h e r country. In isolation, t h e rewards t o labor (wages) a r e lower in t h e country with more labor. When international trade in commodities opens, t h e prices

?f commodities move closer to each other across t h e two regions. As t h e international and domestic prices of t h e two goods become equal, higher levels of welfare a r e attained by both countries because each can consume more by specializing in the exports of t h e good i n which it h a s a relative advantage, while importing a t lower prices t h e good in which the other has an advantage.

These welfare gains are usually referred to as "gains from trade" as, discussed by Samuelson (1962). Figure 1 illustrates these points.

A f u r t h e r result links these gains from trade and specialization with t h e equalization of factor prices. In general, factor prices can be different i n t h e two regions because factors a r e not traded internationally. Stolper, Samuel- son. and Pearce found conditions u n d e r which, following the equalization of commodity prices in the international market equilibrium. factor prices (wages a n d profits) in t h e two regions will also equalize. These conditions require t h a t both countries have t h e same technologies. as well as o t h e r

(10)

Commoditv B ^A key:

( I ' ^'

.

^t

h

labor-lntenslve

,asic aood) = Y = production set

of region

= Z = prices in trade equilibrium

----

= T = prices in isolation

uilibrium consumption

and consumption

l mports of I

Commodity I (capital-intensive industrial good)

PIt%JRE 1 Gains from trade and specialization; output, prices, and welfare in isolation and A t h trade. In Figure 1

Y

is the production possibility frontier of one region, t h e South: it represents the combination of basic goods B and industrial goods I t h a t it could produce. In isolation, t h i s region maximizes welfare on the production possibility s e t Y, reaching t h e level

W1.

When trade is opened prices change and welfare can be maximized over all income available, which now includes export revenues. The new (after trade) budget s e t Z for t h e region is in general larger than

Y ;

therefore a higher level of welfare

W 2

c a n be achieved with free trade. The tangential intersection pf t h e welfare surface

W 1

with t h e production set Y determines t h e output of goods

B

and I in equilibrium before trade, and their relative price is given by t h e tangent line

T.

When international trade is opened, t h e price of t h e labor-intensive good

B

increases, abd a t t h e new prices

S

t h e budget for the country is given by t h e triangular set

2.

Note t h a t after t r a d e basic goods a r e relatively more valuable than before, and more

B

is produced: t h e output of

B

increases from B' to

B ~ .

More labor is employed a t f t h e new prices since

B

is laborintensive. Wages c a n be shown to increase with a n increase in t h e price of B (see the Appendix). The country specializes in and exports

B,

it imports I, and it i s b e t t e r off after trade since i t consumes more of both. In any case, t h e welfare level after trade,

w2,

is larger t h a n t h a t before trade,

wl.

(11)

constraints. Under these conditions, therefore, free trade will lead to a rise in the real wage in the labor-abundant region. Free trade, and specialization in the production of t h e labor-intensive good would therefore lead to an improve- m e n t in the distribution of income within the South, a s well a s a n increase in the welfare of both regions. These results encourage, therefore, t h e international division of labor, in which t h e labor-rich region specializes in labor- intensive goods, and the capital-rich region in capital-intensive goods. Note that, a s wages in the, South increase through free trade, the relative advantage of t h e South al$o decreases as trade proceeds. This theory, therefore, Predicts t h a t relative advantages tend t o disappear with continued trade, and t h a t the international division of labor is a temporary, even self-destructing phenomenon. Whatever inequalities are introduced by it a r e predicted to be purely temporary; in the long r u n the world economy is expected to move towards a more equal state. R g u r e 2 illustrates these points.

The results on gains from trade and specialization and on factor price equalization have been powerful enough to shape most formalized thinking on the theory of trade a n d international economic relations. They have also per- meated policy thinking in a pervasive manner.

However, it is becoming increasingly clear t h a t t h e r e are several factors that have not been considered i n these theories, and which may have a striking effect on market behavior. Moreover, concerns have arisen about the general validity of policy thinking based only on Heckscher-Ohlin theory.

It is generally acknowledged t h a t this theory h a s not provided a n adequate explanation for a salient feature of the postwar period (1945-1970). Dur- ing this period, t h e volume of international trad.e increased in an historically unprecedented fast?ion,' while wealth differences and t h e division of labor between t h e North and t h e South increased significantly.4 Furthermore; t h e distribution of income within t h e S.?uth did n o t improve during this period.5 Neither the results on gains from trade and specialization nor those on factor price equalization seem consistent with these facts. Of course. exogenous historical explanations could be invoked, but this would amount t o an implicit recognition of t h e limited explanatory powers of the theory.

A number of alternative explanations have been proposed for the striking developments in t h e world economy during t h e period 1945-1970. However, these have not produced a body of formalized theory with t h e rigor of t h e neoclassical theory. Therefore, they have lacked policy utility and have n o t carried a s much conviction as the neoclassical theory.

The most significant representatives of these alternatives theories can be grouped according t o the weights t h a t their analysis assigns t o market vs.

nonmarket factors. In general. those t h a t assign more weight to markets t e n d to offer a higher degree of formalization or quantification.

Significant representatives of the analysis t h a t focuses primarily on markets a r e R. Prebisch and W . k Lewis. Both emphasize t h e dangers of over- reliance on international markets as an "engine of growth" for the South.

Prebisch gave a variety of market-related explanations for North-South inequalities, perhaps t h e best known being his work on t h e deterioration of t h e secular t e r m s of trade between products exported by t h e industrial countries and those exported by the "periphery" (Prebisch 1950, 1959). Prebisch's analysis is based on t h e premise t h a t the demand for raw materials a n d labor- intensive products rises less than the demand for industrial products as incomes rise. Thus, over time, the relative prices of products exported by t h e South must decrease. Prebisch's work has led to t h e emergence of what a r e

(12)

Price of exportables B PB

Abroad,

Domestic

Demand from international market

I I I I

Wagelrental ratio Quantity of exports

FIGURE 2 Factor price equalization effects of trade. In Figure 2, the left-hand diagram describes the relationship of the (relative) price of the exportable good

B

with t h e wage-rental ratios w / 7 . Since t h e exportable good is more labor intensive, a s i t s price increases, so does t h e wage-rental ratio. The right-hand diagram shows the possible equilibrium values of exports and imports of the South and the North. In isolation. the equilibrium quantity of exports at home is zero, and the price is t h e domestic price pd. Similarly, the equilibrium price abroad is pa in isolation. When international trade opens, a n international trade equilibrium price p* is reached, and the quantity of exports equals

Q* .

A t price p*

.

the new equilibrium wage rental ratio is (W / T )

*,

which b larger than the isolation wage rental before trade a t home (w/ T ) ~ , and small- e r than the wage rental before trade abroad ( w / T ) ~ . Factor prices and prices of com- moditier equate in the trade equilibrium and improve the wage rental ratio i n t h e South to ( w / T)*.

(13)

now widely known as import-substitution policies in Latin America. Lewis (1952, 1977) has contributed a wide range of powerful insights into t h e economic relationships between t h e North and the South, the most celebrated s e t of ideas emerging from his model of developing economies with unlimited supplies of labor and t h e determination of what he refers to as "factoral t e r m s of trade" for North-South trade. Lewis assumes that labor'supply in the South is infinitely elastic and t h u s t h a t t h e level of employment is determined solely by demand. The real wage is pegged to the subsistence level, a n d t e r m s of trade between the two regions are determined by t h e i r respective levels of labor prdductivity in agriculture. Lewis' analysis leads him to conclude t h a t export-led policies may have a limited value for the South, t h a t t h e growth of the North is not necessarily linked positively with the growth of t h e South, and t h a t there is a need for an endogenous "engine of growth" within t h e South.

The dependencia theory in Latin America, some of whose representatives a r e F. Ca;doso. P. Baran, G. Frank, T. dos Santos, 0. Sunkel, and M.C. Tavares, combined: explanations of underdevelopment based on post-Marxist power relations with a n analysis of markets. These theories had little or no formalization, and therefore could n o t be used for policy f o r m u l a t i ~ n . ~ A c e n t r a l t h e m e in these studies is t h e impact of a "foreign industrialized sector" on the r e s t of t h e economy of a developing country. This foreign sector is associated with tedhnologies. market demand structures, and distributions of income, i.e. with patten& of development, which reinforce the dependence of t h e developing economies on t h e economies of t h e center. Finally, a number of other authors have largely disregarded the behavior of the market a s having insignificant explanatory power, and have addressed themselves instead t o t h e power struc- t u r e of classes and to certain historical developments of capitalism, s u c h as accumulation of capital within and between the industrial and developing regions. Arnong this last group, t h e most significant a r e the Marxist economists k Emmanuel and S. Amin, neither of whom sought t o formalize their assumptions o r results.

Formalized economic theory allows for more accurate empirical testing.

and also for t h e evaluation of alternative policies. Perhaps even more impor- tantly, formalization allows for more consistent "thinking through" of ideas.

Formalized theories can therefore grow, disperse, and frequently be applied better than nonformalized ones.

Another advantage of formalization is t h a t it allows one t o compare, in a precise manner, the assumptions of different theories, so t h a t t h e discrepan- cies in their results can be explored with precision. This can improve t h e level of the analysis and help clarify the validity of the results. With this understanding we decided, within our UNITAR project,7 t o develop a body of formalized theory t h a t could be helpful i n analyzing some of t h e stylized facts of development and trade t h a t appear t o defy explanation by the existing formalized theories. Simplified versions of o u r model have been produced t h a t a r e rather close t o t h e general equilibrium trade models of t h e Heckscher-Ohlin type. Yet, under certain conditions o u r models yield results t h a t appear strik- ingly different from t h e conventional results of gains from t r a d e a n d factor price equalization exercises. This is t h e subject of the next section.

(14)

3.

A

MODEL OF NORITH-SOUTH TRADE

In this section we present a simplified version of t h e North-South model.

The equations, main theoretical properties, and simulations a r e presented in the Appendix. This model differs from the Heckscher-Ohlin model in two main respects: t h e production system of t h e South is different from t h a t of t h e North, and the labor and capital supplies of each region a r e responsive to real wages and profits. By contrast, in the Heckscher-Ohlin model, the two regions have the same production systems and total endowments of factors a r e fixed.

This model was estimated with time series data for Sri Lanka and t h e United Kingdom trading with each other, and t h e estimation confirmed t h e results which we shall discuss next.' Sri Lanka and the United Kingdom were chosen to estimate the parameters of the model because a s trading regions they correspond roughly to t h e stereotypical assumptions of the North-South model. Simulations of the model have also been performed with data for Argentina, Mexico, and t h e United states.'

We shall now describe those characteristics t h a t t h e model has in common with t h a t of Heckscher-Ohlin. There a r e two regions, denoted North and South. Two aggregate goods are produced, consumed, and traded by each region. These a r e denoted B, basic consumption goods, and I, industrial goods.10 Neither region is completely specialized in production of one or t h e other good. There a r e two factors of production, capital and labor, in each region; t h e owners of these factors produce, consume, and t r a d e the two goods. Goods are produced using constant-returns-to-scale technologies. The industrial good is more capital intensive and the basic good more labor intensive in both regions. The South exports basics and the North industrial goods.

In a trade equilibrium, the amounts produced, consumed, and traded within and between the regions a r e determined through the clearing,of mark- e t ~ , i.e. supply equals demand. There a r e four domestic markets (two markets for factors, and two markets for goods) and two international markets (for goods). The international market equilibrium determines t h e prices of both goods and t h e r e t u r n s to factors, i.e. wages and r a t e s of profits in each region.

Each trading region is constrained by i t s budget, so t h a t export revenues and import costs a r e i n balance. In equilibrium the relative price of each traded good is t h e same in the North and i n t h e South.

We now discuss t h e differences between our North-South model and t h e Heckscher-Ohlin model. As already mentioned, t h e two regions have h e r e different technologies for t h e production of goods,11 and the supplies of factors of production in the two regions, capital and labor, a r e responsive to their rewards, r a t e s of profits and wages. Furthermore, we assume t h a t t h e responsiveness of labor supply t o real wages is r a t h e r large in t h e South, while less so in t h e N ~ r t h . ~ The high responsiveness of labor supply t o real wages is a meas- ure of t h e abundance of labor in t h e South. This could be interpreted as indicating a high level of migration from t h e subsistence part of t h e economy (such as rural areas) t o t h e market economy. In a limiting case, when this responsiveness is infinitely large so t h a t labor supplies never represent a con- straint, this would be identical to Lewis' assumption of unlimited supplies of labor. However, Lewis' assumption implies t h a t wages a r e always pegged a t the subsistence level, while in the North-South model, by contrast, wages do vary with supply and demand conditions. This is because despite t h e fact t h a t labor supply is highly responsive t o wages, i t is not infinitely elastic in our model. The high responsiveness of t h e labor supply in the South is denoted abundance

01

labor. We should note here t h a t t h e responsiveness of the s u p ply of a factor t o its price may depend not only on i t s relative scarcity but

(15)

also, in some cases, on the degree of market control exercised by the group t h a t supplies t h e factor (i.e. workers or capitalists). For instance, a t present, labor in t h e South is clearly much less organized, and has less market power than in the North.

The behavior of factor markets is formalized as follows. We assume t h a t labor supply increases with the real wages.

where

w / m

is t h e real wage and a is the positive response of labor supply to real wage. Similarly, for capital supply

K = ~ T + K ,

p > o

where r is the r a t e of profit and /3 is a positive response parameter. In t h e South the parameter a is large, i.e. labor supply is very responsive to real wages, and the parameter

/3

is small, i.e. capital supply is not too responsive t o the r a t e of return. The opposite is t r u e of the parameters a and

/3

i n t h e North. Special cases of this model, where

K

= and L

= i,

have been studied (see chichilnisky 1983a).

We now discuss t h e results on gains from trade and specialization a n d on factor price equalization in t h e context of t h e North-South model, and we shall explain h o y t h e characteristics of this model lead to t h e various results. We s t a r t by indicating, by means of Figure 3, how the geometric illustration of t h e gains from t r a d e ?nd specialization, given in Flgure 1, is nu l o n g e ~ universally valid w h e n factoraare in variable supply.

The second differende between the present model and t h e Heckscher- Ohlin model is t h a t in the North-Sputh model different regions have different technologies. This is formalized by assuming different production functions lor t h e two regions.

In t h e Heckscher-Ohlin model, production functions are assumed t o show constant r e t u r n s to scale, and t o be identical in both countries. We also consider constant-returns production functions, though for analytical simplicity we use h e r e fixed factor proportions. Extensions t o Cobb-Douglas or

CES

production functions a r e discussed in Mahran (1982). Let B denote a basic good, and I an industrial good. If B' and denote outputs of the two goods, t h e two productioli functions a r e

BS

=

rnin (L B/ a I ,

KB/,

IS

=

min (L'/ a2,

K'/

c 2)

where c , and c z a r e capital/output ratios a n d a l and a2 a r e labor/output ratios in t h e two sectors, respectively. As is well known for such technologies.

the difference in factor intensities in t h e two sectors is given by t h e elrpres- sion D

=

a 1 c - a z c We ossume here that D is positive in both regions, but much larger in the South t h a n in the North. This means that the good

B

is produced in a more labor-intensive way i n both countries than the good I, but B is much more labor-intensive than I in the South.

Casual obsemation suggests t h a t differences in capital/labor intensities between t h e technologies of the two sectors a r e indeed larger in t h e South than in the North. Empirical observations confirm this. Developing economies have significant differences b e b e e n the production systems of

(16)

S = prices after trade

,

⁼prices before trade

a n e r rrauu

IL

..

I '

Commodity I (capital-intensive good)

FIGURE 3 Gains from trade with price-responsive endowments. Figure 3 differs from Figure 1 in t h a t initial endowments of capital and labor vary h e r e with prices so that : t h e production sets also vary with prices. The pre-trade production possibility set is

yl. The equilibrium relative prices before trade are given by t h e line

T1;

t h e equilibrium quantities of the two goods produced pre-trade are

B 1

and

Il.

After trade, t h e production possibility frontier has changed due to the response of the variable factor supplies; it is now

Y2.

This could not occur in Figure 1 since factor supplies were fixed there. Labor supply has now increased with respect to capital goods supply, and therefore the new production possibility set

Y2

shows that the economy is now able to produce relatively more labor-intensive goods B. New prices S a r e achieved i n equilibrium that reflect an increase in the price of the capital-intensive good I. The output of the labor-intensive good has increased with respect to that of t h e capital-intensive good (to

B2

and 12) but total welfare levels achievable at t h e equilibrium

(W2)

a r e now lower. The figure illustrates how a lower level of welfare is possible after trade, if factor endowments a r e price dependent.

(17)

different sectors, and we refer t o this as "technological dualism". This fact has been pointed out in the theories of dual economies for many years, starting with the less formalized work of t h e Latin American economists already mentioned. However, a major difference between our work and previous studies of dual economies is t h a t while those studies attempted to explain a division of t h e economy into a market and a subsistence sector, here we consider t h a t markets operate throughout the economy, and dualism appears here only in production. Otherwise, in our model, the economy is completely integrated and all its parts interact in a general equilibrium with each other. In particular, wages and prices a r e determined through the interaction of all markets.

I t should therefore be kept in mind t h a t the t e r m s "dual technologies" or

"dualism in production" have r a t h e r different meanings here than elsewhere in the literature.

4. BASIC NEEDS

AND NORTH-SOUTH TRADE

We now summarize t h e results obtained with the North-South model.

When the economies of t h e South have very abundant labor and significant dualism in production, an increase in t h e exports of t h e labor-intensive basic consumption good will necessarily decrease the price of this good in relation t o t h a t of the industrial good; domestic employment a n d t h e purchasing power of wages will also decrease. The domestic consumption of basic goods decreases as well. This will occur quite independently of t h e cause of this increase in exports; for example, i t may occur even with an expansion of t h e North's demand for exports from t h e South. In particular, the results do not depend on any assumptions of t h e elasticity of international demand for goods from the Sbuth. They depend instead on domestic conditions in the South.

The specific condition is t h a t a be large and t h a t c 2 / D

<

Zw/ PB, where w / PB is the real wage. (Note t h a t when D is large, technologies are dual and c 2 / D is more likely to be smaller than, r a t h e r than twice as great a s t h e real wage.) This second condition is termed "dualism".12

What these results show is t h a t under conditions of dualism and abundant labor in the South, a higher volume of exports is necessarily associated with a loyer (relative) price of t h e basic good, with lower wages and employment, and with lower domestic consumption of basic goods. In the following, we summarize t h e rationale for this sequence of events. However, before doing so i t seems worth noting t h a t t h e results a r e r e v e ~ s e d when t h e production system in the South is more homogeneous and when its labor supply is less abundant.

In this latter case, following a n expansion of exports, we obtain results in t h e spirit of "gains from trade and specialization": increasing exports leads to an improvement in the North-South t e r m s of trade, and t o increases in domestic employment, consumption of basics, and real wages. From an economic viewpoint, therefore, t h e economic parameters of dualism and of labor abundance a r e rather important. Such parameters may have t o be modified before embarking on an export-led policy. It should be noted t h a t t h e crucial duality condition c z / D

<

2w/ PB t h a t determines whether or not an export expansion is desirable can be shown t o depend itself on the level of exports already achieved by the economy. Therefore, different export policies a r e advisable a t different export levels. The results may therefore be viewed a s suggesting optimal export levels, or optimal balances between domestic sectors a n d t h e export sector. We shall discuss these and other alternative policy issues in the last section of the paper.

(18)

We now discuss how t h e negative impacts of increased exports emerge when t h e economy of t h e South has abundant labor and dual technologies.

This requires t h a t we analyze t h e changes in the domestic consumption of basic goods as their price increases.

If all prices other t h a n those for basics remain constant, t h e economist's ceteris paribus assumption, an increase in t h e price of basic goods will lead to lower demand a n d t o lower consumption of basics. This is a standard partial equilibrium demand response to a price change. However, things a r e quite different when we remove t h e ceteris paribus assumption a n d l e t all markets adjust, ite. when we move from one full market equilibrium t o another. In this latter case, t h e demand for B may increase a t a new equilibrium with a higher price of

B.

This will occur when, a t a new price level, firms produce more of the labor-intensive commodity B and increase, therefore, t h e levels of employment and of wage income, thus leading to a higher d e m a n d for t h e basic good a t t h e new equilibrium. Similarly, t h e supply of basic goods increases with a higher price of basics. What remains t o be determined is whether supply has increased more t h a n demand, o r vice versa, a t t h e new equilibrium, Since exports a r e t h e difference between domestic supplies and domestic demand, this determines whether exports increase or decrease a t a higher price of basics. The sign of the expression cZ/ D - 2 w / p B compares precisely the strength of t h e supply ( C ~ / D ) with t h e demand response (2w PB)-

Proposition 1 of Chichilnisky (1981) proves t h a t with abundant labor and dual technologies, t h e increase in domestic demand ezceeds t h e increase in supply a t t h e new equilibrium. Therefore, a s t h e price ;of basics increases, exports decrease. Equivalently, exports can increase -only if prices a n d domestic coqsumption drop. -For a proof, see the Appendix. A c l e a r exposition and technical analysis of this result is offered in Arrow (1981), Sen (1981). a n d Heal a n d ~ c ~ e o d (1983). Figure 5 illustrates t h e behavior of a reaction curve of t h e South, a curve t h a t links t h e volume of exports with t h e price of t h e s e exports across different possible equilibria of this economy. This curve t r a c e s the intersections of domestic supply and demand curves. Note t h a t since this reaction curve violates ceteris paribus assumptions (as all m a r k e t s adjust in it) t h i s curve is not a supply curve for t h e economy. I t merely allows u s t o t r a c e out bypothetical equilibria, a n d t h e relationship between t h e price of basic goods arld t h e volume of exports of basic goods, across e a c h of these equilibria. For each level of international demand, only one equilibrium is possible, the one where t h e volume of exports equals t h e volume of imports demanded by t h e North. This volume of exports corresponds to one price level for basic goods. We c a n therefore analyze t h e changes in prices of basics a s the volume of international exports varies across equilibria. However, a t a n y equilibrium where labor i s abundant and technologies a r e dual ( a l a r g e , c 2 / D

<

2 w / p B ) , a move towards another nearby equilibrium with a higher volume of exports from t h e South is necessarily associated with a lower price of basic goods.

The price of the basic good is always positively associated with wages because the basic good is labor intensive (see the Appendix). I t follows t h e r e - fore t h a t as exports increase a n d the price of the basic good decreases, so do wages. The purchasing power of wages in t e r m s of t h e basic good is also shown t o decrease when exports i n c r e a s e (see t h e Appendix). We therefore conclude t h a t u n d e r dualism a d with abundant 1abo.r (a large ,c 2/

D <

2w /pB), a move towards a nevl equilibrium, with a high67 volwne of ezports is necessarily asso-

(19)

c i a t e d w i t h a l o w e r price of these e z p o r t s , a l o w e r r e a l w a g e , and d e c r e a s e d d o m e s t i c c o n s u m p t i o n of basic goods in the South. This is Proposition 1 in Chi- chilnisky (1981).

I t is of i n t e r e s t t o point out t h a t should t h e economy's technologies be more homogeneous, and labor supply less responsive t o wages, these results would be reversed. In this latter case, an increase in t h e volume of exports is associated with a higher level of prices of basic goods. This is because now, a s the price of t h e basic good increases, domestic supply increases proportionately m o r e t h a n domestic demand in t h e new equilibrium. M t h a b u n d a n t l a b o r a n d h o m o g e n e o u s technologies ( c z / D

>

2 w / p g ) a n i n c r e a s e in e x p o r t s l e a d s to b e t t e r t e r m s of t r a d e , to higher r e a l w a g e s a n d e m p l o y m e n t , a n d also t o h i g h e r l e v e l s of c o n s u m p t i o n of basic goods. This is Proposition 2 in Chichil- nisky (1981). It can also be shown t h a t the volume of industrial goods imported also increases in t h e new equilibrium. This is discussed in t h e Appen- dix. Figure 4 illustrates t h e standard case in which, a s exports of t h e basic good expands, t h e price of this good increases, a n d so do real wages and employment. Figure 5 illustrates t h e case of t h e dual economy with abundant labor: a s exports of t h e basic good increase, t h e price of this good, t h e real wage an6 t h e employment all decrease in the South.

A few analogies and contrasts can be drawn h e r e with t h e results of t h e Heckscher-Ohlin theory presented in t h e previous section.

A first point is t h a t , in the North-South model, t h e international division of labor is actually r e i n f o r c e d by t h e expansion of international t r a d e when technologies a r e dual and labor abundant in t h e South. This is because 'more trade leads t o lower wages in t h e South, therefore reinforcing t h e relative advantage of t h e South. In contrast t o t h e conventional wisdom, inequality and t h e division of labor may perpetuate themselves.

A second point is t h a t factor prices not only fail t o equalize across regions in t h i s model, but t h e y also t e n d t o drift further a p a r t in the South a s t r a d e expands. This point is worth mentioning because i t is only n a t u r a l t h a t with different technologies in t h e two regions, factor prices would never become fully equalized. As only goods are traded internationally in this case, only t h e prices of goods a r e equal in equilibrium. However, in a Heckscher-Ohlin model in which technologies a r e different, but duality is n o t significant and labor is not abundant. free t r a d e would f e n d t o equalize (even though i t may never fuLLy equalize) factor prices. Thus in such a Heckscher-Ohlin model relative advantages t e n d t o be blurred as trade takes place, a n d t h e division of labor appears again t o be a temporary or self-destructive phenomenon. In t h e North-South model, by contrast, relative advantages a r e emphasized through free trade because wages in t h e South decrease. The international division of labor is therefore self-perpetuating.

I t now remains t o analyze t h e response of total export revenues of t h e South. This is relevant because if total revenues did increase, i t would be possible t o increase exports and t o redistribute t h e proceeds adequately in order to compensate for t h e harmful effects of export-led policies on wages an( consumption. In Chichilnisky (1983a) we examined how a n increase in exports affects export revenues of t h e South under different conditions, and showed t h a t , with duality and abundant labor, t h e deterioration in t h e t e r m s of t r a d e of t h e South following an increase in exports leads also to lower export revenues. ht the n e w e q u i l i b r i u m the North i m p o r t s m o r e basic goods a t a Lower p r i c e , and e z p o ~ t s a l o w e r v o l u m e of i n d u s t r i a l goods in e z c h a n g e , h c e r t a i n

c a s e s the N o r t h m a y actualLy i n c r e a s e its d o m e s t i c c o n s u m p t i o n of b o t h goods, basic a n d i n d u s t r i a l , af t h e nsw e q u i l i b r i u m .

(20)

Price of B

a. Domestic economy of South ^b.International economy

World excess demand at p i * World excess

. * *

Quantity of B Quantity of B

/ \

Domestic supply Domestic demand across equilibria across equilibria

HGURE

4 The standard case with homogeneous technologies: c 2 /

D _>

2 w / p B . Figure 4a represents the domestic: market, both supply and demand for t h e exportable a t home; Figure 4b represents t h e international market. The difference between t h e equilibrium level of supply and t h e equilibrium level of demand is denoted EB*, t h e amount exported to t h e international market a t equilibrium p r i c e s p B L . Domestically we have t h e corresponding equilibrium values of supply a n d demand t h a t determine domestic consumption a t th.e world equilibrium.

In t h e second equilibrium t h e r e a r e more exports

IT',**

a t a higher price p B 8 * , and a n increase i n t h e equilibrium value of domestic supply, a s well a s a decrease i n equilibrium demand. Therefore, as shown earlier in Figure 2 (left-hand side), t h e wage/rental ratio has increased in t h e South. More exports lead to a higher level of wages i n t h e new equilibrium.

Excess supply

I

^II ^arPs

from South I

at.; I I 1 I I I I I I I

I I

(21)

Price of 6

a. Domestic economy of South

pBl

6. International economy World excess

Excess supply Domestic supply

across

- - -

across equilibria

4

E;, E;*

/ . I

World excess

Domestic supply across

orne nest

ic demand ' Wo

across equilibria

I

^{demand at} equilibria

4

E;, E;*

Quantity of 6 Quantity of B

FIGURE 5 The dual economy with abundant labor: c Z /

D <

2 w / p B . Figure 5 represents domestic and international markets interacting with each other; t h e domestic market is represented in Figure 5a and the international market in Figure 5b. Here pB* represents first equilibrium prices, and EB* t h e equilibrium level of exports. pB** and

EB**

represent the second equilibrium prices and export volumes.

The rest of t h e symbols correspond to those of Figure 4. A s opposed to the neoclassical case, in the dual economy t h e reaction curve that links export levels with equilibrium prices is downward sloping, reflecting the supply and demand effects across equilibria that were discussed in the text for both markets for goods and for services. This is proved in Proposition 1 of Chichilnisky (1981). With increased exports, the wage/rental ratio decreases, and there is also lower total output, and t h u s employment domestically. Furthermore, t h e increase in demand from t h e North brings about lower equilibrium prices for the exportabie.

(22)

In the S u t h , the w o r s e n i n g of t h e t e r n of t r a d e l e a d s to f e w e r i m p o r t s o f i n d u s t r i a l g o o d s . l h e c o n s u m p t i o n of b a s k g o o d s d e c r e a s e s in t h e S u t h , a n d t h e v o l u m e of i m p o r t e d i n d u s t r i a l goods a v a i l a b l e f o r i n v e s t m e n t d e c r e a s e s a l s o . h this s e n s e e z p o r t - l e d g r o w t h u n d e r t h e c o n d i t i o n s of t e c h n o l o g i c a l d u a l i t y a n d a b u n d a n t l a b o r m a y h a v e a h a r m f u l i m p a c t o n i n d u s t r i a l i z a t i o n as w e l l as o n t h e s a t i s f a c t i o n of b a s i c n e e d s . A c o n t i n u a t i o n of e x p o r t - l e d poli- c i e s u n d e r t h e s e c o n d i t i o n s d o e s n o t s e e m p r o p i t i o u s f o r m t a i n e d d e v e l o p - m e n t . All these results o c c u r in stable markets. They a r e proved in Chichilni- sky (1981, 1983a).

With abundant labor and dual technologies, t h e international m a r k e t appears, therefore, to c o n c e n t r a t e r a t h e r than to diffuse t h e gains from trade.

It is t r u e t h a t both regions a r e better off trading t h a n t h e y would be in isolation. But starting from one free trade position and moving to another with a higher volume of exports is not desirable for t h e South when t h e r e is abundant labor and dual technologies, c Z / D

<

2 w / p B . An expansion of exports under these conditions may lower t h e welfare of the South a n d increase t h e welfare of t h e North in a sustained fashion. Obviously we c a n n o t apply these results t o compare autarky and free trade, since autarky is n o t a free t r a d e equilibrium and h e r e we a r e only comparing free trade equilibria with different t r a d e levels. In cases where a u t a r k y is a limit of a sequence of free t r a d e equilibria, i t can be shown t h a t t h e crucial condition c2/

D

- 2 u r / p B becomes positive since the real wage w / p B goes to zero. This means t h a t in these particular cases, more trade is initially belieficial but becomes more harmful as i t expands.

Note, however, t h a t when t h e r e is a minimum wage w / p B required for sub- sistence, t h e r e may be n o export level a t which c 2 / D exceeds 2 w / p B .

The above results s e e m more consistent than t h e standard Heckscher- Ohlin results with t h e bverall international experience of the 1945-1970 period, a period in which t h e r e was an expansion of trade, accompanied by continued specialization and increasing wealth differentials between t h e industrial a n d t h e developing countries. It remains to evaluate t h e generality of t h e assumptions, both within t h e context of what is usually assumed in t h e body of economic theory and also in t h e context of t h e empirical d a t a available, and this we shall discuss next.

The condition of duality in technologies of t h e South is r a t h e r general a n d h a s been tested for Sri Ldnka, Mexico, and Argentina. However, t h e condition of labor abundance (a large) may be considered m o r e stringent. For this rea- son we have also considered a version of th'e North-South model where labor i s not necessarily very responsive t o wages in t h e South, i.e. a may be small o r even zero (see Chichilnisky 1983a). Here we make instead a different assump- tibn, namely t h a t t h e exported good is a "wage g o o d , i.e. a good t h a t coristi- t u t e s t h e bulk of t h e consumption of wage earners ( w L

=

p B B in equilibrium).

This assumption a c t s a s a substitute for t h e condition of labor abundance in producing t h e results. Therefore, an economy with dual technologies a n d which exports wage goods is subject to t h e same harmful effects of export-led policies. Other extensions of t h e model were produced, leading t o similar results, for cases where t h e country exports a labor-intensive good t,hat i s n o t consumed domestically (an export enclave), or where t h e country exports raw materials whose production is not labor intensive; s e e Chichilnisky (1981, 1982). I t seems necessary, therefore, t o correct t h e conditions t h a t lead t o negative outcomes before undert.aking an export-led policy. If t h e conditions a r e intrinsically difficult t o correct, it seems desirable t o consider alternatives t o export-led policies on labor-intensive products. This will be discussed in t h e last section.

(23)

We shall now discuss briefly the relationship of the assumptions a n d results of this model with t h e work of R. Prebisch and of W.k Lewis mentioned earlier.

Prebisch (1950, 1959) developed a thesis t h a t t h e r e is a systematic bias in the distribution of t h e gains from trade against developing countries, implied by a secular deterioration in t h e t e r m s of trade of t h e South. While no formal model was presented, a n economic basis for this process can be summarized as follows. Prebisch postulates t h a t the income elasticity of international demand for exports from t h e South is low, while t h e demand for exports from the North is highly income elastic. Increases in income t h u s proportionately reduce the demand for exports from t h e South, but increase t h e demand for exports from t h e North. This leads t o a secular decrease in t h e price of exports of the South with respect' to t h e exports of t h e North. Other argu- ments were also advanced about the role of noncompetitive agents, such a s large corporations and unions in t h e North, in t h e decline in t h e relative price of exports from the South, and in the rise in income by t h e North. An important outcome of t h e Prebisch "terms-of-trade" thesis were t h e protectionist policies of import substitution in Latin America in t h e l a t e fifties a n d t h e early sixties. These policies imposed tariffs in order t o protect c e r t a i n domestic industrial sectors, such a s manufactures and capital goods--the so-called infant industries.

Our approach is different from t h a t of Prebisch both in assumptions a n d in results. In t h e first place, the North-South model makes no assumptions about the elasticities of international demands for t h e goods exported by t h e North and the South. Our assumptions a r e instead on d o m e s t i c s t r u c t u r e s : technologies a n d factor m a r k e t s within each region. Secondly, o u r model i s consistent with perfectly competitive markets for goods a n d factors and, therefofe, also differs from t h e assumption of Prebisch about noncompetitive agents in the North. Finally, with respect to policy: certainly t h e North-South model does not advocate t h e replacement of export-led growth by import substitution policies. This is because import substitution is a policy concerned only with t h e supply side of t h e economy, while in o u r work, instead, both supply and demand must be considered t o evaluate t h e outcomes. Appropriate local demand s t r u c t u r e s s e e m a t least as important a s changes in t h e supply side, t o obtain beneficial outcomes.

We focus next on W.A. Lewis' celebrated model of economic development with unlimited supplies of labor (Lewis 1952), and especially on his l a s t section where he discusses development and trade. Lewis' model considers two regions trading with e a c h other. One, t h e North, has t h e characteristics of a neoclassical economy. The o t h e r region, the South, is characterized by unlimited supplies of labor a n d a dual economy, a p a r t of which is capitalistic a n d the o t h e r "traditional." Lewis' work, therefore, diverges from t h e neoclassical model as well a s from o u r model, in t h a t a r a t h e r different formalization i s given t o the economies of t h e North a n d of t h e South. The North is a perfect market economy, while t h e South has a r a t h e r different s t r u c t u r e . In o u r model, instead. t h e behavior of both regions is consistent with perfect m a r k e t behavior.

Clearly, t h e assumption of a high elasticity of labor supply in o u r model i s linked with Lewis' assumption of unlimited supplies of labor. However, o u r assumption is substantially different: wages in t h e South do adjust with changes in market conditions. In Lewis' model, instead, wages a r e per- manently pegged t o t h e subsistence level. Another difference i s t h a t Lewis' model assumes t h a t t h e economy i s divided i n t o a "capitalist" a n d a

(24)

"traditional" sector. In t h e capitalist sector, the motive for employment is t o g e n e r a t e profits, while in the traditional sector, labor is considered as essentially self-employed ( a s in t h e peasant family) or engaged in petty trade, or service occupations. The real wage in t h e capitalist sector is .exogenously given, and i t exceeds earnings available in the traditional sector, .so t h a t employment in the formal sector is constrained by demand only and not by supply. I t is in t h i s sense t h a t Lewis speaks of "unlimited supplies of labor".

Given the wage a n d technology, profit maximization determines t h e capital/labor ratio a n d t h e r a t e of profit, and the size of t h e capital stock determines t h e level of employment in t h e capitalist sector. Each region produces t h r e e goods, one of which is common t o both. The t e r m s of t r a d e between t h e two regions a r e determined purely by relative labor productivities in t h e common good, food, independently of demand conditions. Demand conditions a r e therefore not important in t h e Lewis model. All t h e s e specifica- tions c o n t r a s t with t h e North-South model.

In t h e North-South model, instead, demand is r a t h e r important. I t helps determine t h e variable level of real wages and it contributes t o the understanding of t h e relationship between international m a r k e t behavior (export levels) and domestic output and distribution of income. The international t e r m s of trade a r e determined by supply and demand forces in both markets.

I t is also of interest to c o n t r a s t our model and the Heckscher-Ohlin model with t h a t ~f Lewis. Lewis considers two different seckors, a capitalistic and a traditional one, each with different modes of behavipr. Only the capitalistic s e c t o r maximizes profits. Instead, in our economy, m a r k e t behavior is con- s i s t e n t with t h e hypothesis of profit a n d utility maximization throughout. As in t h e Heckscher-Ohlin model, both economies produce, consume, and t r a d e t h e s a m e two goods, while Lewis considers t h r e e goods in each region, only one of which is s h a r e d by both regions (food). These features make Lewis' model difficult t o compare directly with t h e two-region, two-good, a n d two- factor Heckscher-Ohlin model, and t h u s his results, while yielding different conclusions, do not necessarily contradict or support t h e standard results. By contrast, since our r e s u l t s a r e posed in a manner completely analogous t o t h a t of t h e Heckscher-Ohlin model. a more thorough comparison of assumptions a n d results is possible, a n d t h u s criticism can be formulated more precisely.

Finally, we compare t h e North-South model with t h e work of t h e Marxist economists a n d t h e dependencia theorists of Latin America. Both t h e s e groups give less importance t o market behavior than we do. G. Frank concen- t r a t e s on t h e secular t r e n d s in t h e international accumulation of capital a s determining t e r m s of trade, while o u r work does n o t produce results t h a t predict a n y secular t r e n d s in this sense. However, if the labor m a r k e t behavior t h a t we s t u d y i n our model could be related t o the stages of t h e accumulation of capital of t h e South, t h e n the deterioration in t e r m s of trade t h a t h e predicts could be, in part, attributed t o market forces. A Emmanuel's unequal exchange work measures t h e t e r m s of trade by t h e value of a u n i t of labor in t h e North relative t o t h a t of a unit of t h e South's labor. Trade between economies with different wage levels, in his conceptual s c h e m e , m u s t r e s u l t in "exploitation" of t h e one with the lower wage. This view is, in c e r t a i n cases. consistent with t h a t of Lewis on factorial t e r m s of trade. In Emmanuel's work ( a s in Lewis') t h e real wage is exogenously fixed i n each c o u n t r y while. a s explained above, in our model real wages adjust i z relation t o t h e international t e r m s of t r a d e and to all markets. This is a significant f e a t u r e of o u r results.

North-South Trade and Basic Needs