University of Southern California Dissertations and Theses

Essays on some Lewisian themes of economic development

(USC Thesis Other)

Essays on some Lewisian themes of economic development

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ESSAYS ON SOME LEWISIAN THEMES OF ECONOMIC DEVELOPMENT by Anup Kumar Sinha A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Economics) September 198 3 UMI Number: DP23325 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Dissertation Publishing UMI DP23325 Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA 9 0 0 0 7 '33 PW.D- to This dissertation, written by ...... Anup_ Kumar Sinha under the direction of h.X-3... Dissertation C o m mittee, and approved by a ll its members, has been presented to and accepted by T h e G raduate School, in p a rtia l fu lfillm en t of requirements of the degree of D O C T O R O F P H I L O S O P H Y .................( y P 'Dun D a t e . . . . ^ k » J ^ J ^ l l . . . ^J2ISSEi^TION_CpMM ITTEE^-j— ... 0 — 0 o. JChairman 2w ± .... J>96>3F v? ACKNOWLEDGMENTS I would like to thank and express my deep gratitude to Professor Jeffrey Nugent for his constant encouragement. He gave me, perhaps, more than a fair share of his time from an extremely busy schedule. I consider it an honor and a privilege to have been his student. I would also like to thank Professors John Elliott and Eugene Cooper for their useful suggestions and the interest they showed in my work. I am indebted to Professor Samar Datta for long and stimulating discussions. My colleagues, Bernardo Gonzalez-Arechiga, Abdelhamid Mahbob and Subal Kumbhakar helped me with critical comments and suggestions. The economics department at USC provided a friendly and helpful atmosphere to smooth Out the ups and downs of graduate work. A heavy academic schedule did not prevent my wife, Madhumita, from bearing the lion's share of household responsibilities, with a smile and loving care. Thanking someone who is so near seems so inadequate. Amidst poverty and hardships, my parents struggled to give me the best of opportunities. There is no wealth that will suffice for the repayment of this debt. ii TABLE OF CONTENTS Page ACKNOWLEDGMENTS i i LIST OF FIGURES....................................... V CHAPTER 1 INTRODUCTION.................................. 1 1.1 The Lewis Setting....... .............. 1 1.2 Purpose and Scope..................... 5 1.3 The Problem of the Terms of Trade.... 9 1.4 Technological Change in Agriculture.. 13 1.5 The Issue of Foreign Investment..... 15 1.6 The Outcome of International Trade... 16 1.7 Concluding Remarks............... 18 2 THE LEWIS MODEL WITH FOOD SUPPLY........... 20 2.1 Introduction........................... 2 0 2.2 The Model......... 22 2.3 Comparative Statics.............. 33 2.4 Concluding Remarks.................... 36 Appendix 2.1................. 40 Appendix 2.2............................ 49 3 TECHNOLOGICAL CHANGE IN BACKWARD AGRICULTURE................................... 51 3 .1 Introduction.......... 51 3.2 The Bhaduri-Newbery Debate........... 55 3.3 New Technology under Sharecropping... 60 3.4 Concluding Remarks..................... 6 5 4 TRANSACTION COSTS, FOREIGN INVESTMENT AND ECONOMIC GROWTH......... 67 4.1 Introduction........ 67 4.2 A Rationale for Foreign Investment... 70 4.3 Growth in the Presence of Foreign Investment............. 73 4.4 Policy Implications...... 90 4. 5 Concluding Remarks. . . ................ 92 iii Page 5 INTERNATIONAL TRADE AND GROWTH IN A DUAL ECONOMY........................ 94 5.1 Introduction............................ 94 5.2 The Model........ 98 5.3 The Dynamics............................ 103 5.4 Concluding Remarks..................... 110 Appendix 5.1......... 113 Appendix 5.2............................ 116 6 CONCLUSION.................................... 119 BIBLIOGRAPHY.......................... 125 iv LIST OF FIGURES FIGURE Page 2.1 Excess Supply.................................. 30 2.2 Excess Demand........ 31 2.3 Full Equilibrium.................... 32 4.1 Transaction Costs and the Growth Rate...... 71 4.2 The Rate of Profit on Foreign and Domestic Capital......................................... 75 4.3 kF Has a Positive Intercept and a Negative Slope......... 86 4.4 k„ Has a Positive Intercept and a Positive r Slope.......................... 87 4.5 k„ Has a Negative Intercept and a Negative r Slope.................. 88 4.6 k„ Has a Positive Slope and a Negative r Intercept. ..... 89 5.1(a) Strong Dependency Effect...................... 105 5.1(b) Strong Dependency Effect - The Krugman Case............. 107 5.2 Dominant Internal Effect...................... 108 5.3(a) The Ultimate Domination of the.Internal Effect........................................... 109 5.3(b) The Ultimate Domination of the Dependency Effect........ 110 v CHAPTER 1 INTRODUCTION 1.1 The Lewis Setting Since the publication of W.A. Lewis' (1954) article, development economics has witnessed a rapid growth in the literature on "dual" economic development. Approaches to the Lewis model have been numerous, probably more numerous than any other model. The reactions have varied widely. On the one hand, some authors have extended and formalized the model and used it to study a wide variety of different issues in the economics of development. They have ranged from the exact mechanics of transition from dualism to modernization [Fei and Ranis (1964), Jorgenson (1961)], to problems of economic planning [Bose (1968), Dixit (1968), Hornby (1968)], and the issue of steady- stage growth [Araano (1980), McIntosh (1975)]. On the other hand, there have been a large number of critics who have expressed doubts about the usefulness of the Lewis framework for the study of economic development. The perspectives of these critics have also differed consider ably, from those based on a neoclassical viewpoint, to those based on neo-Marxism. However, in spite of persis tent criticisms, the Lewis model has remained at or near 1 the center stage of the literature on economic development. Before reviewing some of the major criticisms, it is worth while to briefly summarize the main arguments in Lewis' original paper. Lewis essentially drew on the historical experience of the already developed economies, to suggest a general outline, or scheme, of the development process. A capitalist sector leads the economy by drawing on a cheap reservoir of labor from a subsistence sector. Laborers in the subsistence sector are "surplus" in the sense that their withdrawal does not lead to a decline in the subsis tence sector output. In this type of economy, the opportunity cost of an additional laborer transferred to the capitalist sector remains constant. As a result, capitalist sector employers face an infinitely elastic supply of labor. The wage-rate offered to the capitalist sector workers must compensate them both for their loss of subsistence sector income and the direct costs of migration, if any, and beyond that must include a small psychological markup to induce them to migrate. As the capitalist sector expands, the share of profits in national income will rise. Since savings out of profit income is substantial, the savings- income ratio will rise, leading to rapid capital accumulation. When all the surplus labor has been absorbed, wages will rise, and the initial phase of 2 development will come to an end. The crux of Lewis' argument was simple, yet powerful. He pointed out how a cheap and abundant natural resource can be fruitfully utilized to step up the tempo of growth. Lewis was aware that he was not prescribing an easy "how-to-do-it" formula for development. He was merely pointing out a possibility. The caveats he added in his paper indicate this. Even when surplus labor existed, certain problems could arise that would tend to reduce the rapidity of the growth process. First, workers could, by trade unions or otherwise, exert an upward pressure on wage rate in the capitalist sector. Secondly, rising living standards in the subsistence sector could affect the opportunity cost of capitalist sector employment. Finally, if the capitalist sector depended on the subsistence sector for food, then this had to be obtained at stable prices so as not to turn the terms of trade against the capitalist sector. One would have expected that the Lewis model would invite further research in terms of providing more structure and behavioral content, and become the basis of a full-fledged paradigm. Indeed, the initial responses of Fei and Ranis were along those lines. However, in the face of mounting criticism, the Lewis model never became fully articulated into a school of thought. 3 It is not the purpose here to present a comprehensive review of all of Lewis' critics. However, it is deemed relevant to at least provide the gist of two alternative critiques which will subsenquently be examined in greater detail. One such critique has developed along neoclassical lines where dualism is seen to be a temporary aberration. According to this view dualism is the result of market imperfections and distortions. Allowing the market to efficiently allocate resources would make dualism disappear. The automatic equilibrating mechanism of the market would ensure harmonious development through marginal adjustments. In contrast to the above argument, structuralists and neo-Marxists have noted that dualism is the result of the uneven nature of world capitalist development. It is not seen as a temporary aberration; attempts to rely on the market mechanism, moreover, would aggravate the problem. Slow and hesitant development is seen as a persisting feature of a dualistic economy. Unlike Lewis, for them the only solution would lie in a thorough-going social revolution in property rights and institutions. Before examining the criticisms further, a few words are in order regarding the purpose and scope of this work. 4 1.2 Purpose and Scope The purpose of the four essays presented below is to rigorously examine the Lewis setting in the light of some of the criticisms made against it. The task also has a practical significance. The problem of economic develop ment is an urgent one for the less developed nations. Given this urgency, the increasingly divergent literature on economic development makes policy choices difficult. If some of the divergences can be narrowed down and suf ficient conditions identified for the different possibili ties, it can contribute towards a clarification of the options faced by policy makers and planners. The increasing divergence can be observed by classify ing the literature into two stylized categories. On the one hand, mainstream neoclassical economists have main-. ~ tained that increasing the efficiency of the market economy by removing distortions, making the latest technology available and opening up the economy to free trade will facilitate the development process. On the other hand, more radical economists have argued that it is precisely the reliance on a market economy, foreign technology and free trade that constrains; the full potential for ecpnomic development. The differences in these two broad categories of the literature are attributed to dissimilar premises or 5 non-identical preanalytical visions. The purpose of these essays is to show that neither of the outcomes indicated by the two schools of thought are inevitable. They are alternative possibilities in the development process. If these alternatives can be shown as possibilities that can arise in the development process of a Lewisian economy, then the Lewis model can lay a claim to generality. It is in this context that the following essays contribute, in a modest fashion, to a more general theory of economic development. The scope of the essays are set by trying to identify the most important issues of economic development in a Lewisian economy. Within the basic Lewis setting it is crucial to investigate the nature of the interdependence of the two sectors in determining the relative price of the two goods and the implications for industrial employment. Once the basic interdependence is studied, attention has to be turned to each of the sectors separately. If the agricultural sector has to keep pace with the expansion of the industrial sector it must provide more workers and more food to the industrial sector. This was a major proposition of Lewis. The ability of the agricultural sector to release more food and more workers is closely dependent on productivity - increasing technological change. It is important to analyze the 6 problems involved. Is the availability of new technology sufficient to achieve the required increase in productivity or do institutions need to be changed before a new technology will be adopted? The next question that arises is: how can capital accumulation be stepped up in the industrial sector? In this respect there are two closely related issues. One is the need to use more productive technology and the other is to expand the volume of investment. In a Lewisian economy both the agricultural and industrial sectors must find ways to raise productivities. For the industrial sector the need to acquire better technology and higher investments can both be provided by foreign investment. However, there is strong disagreement as to the benefits and costs of foreign investment. The conditions under which foreign investment can act as a catalyst or alterna tively, under which foreign investment may come to dominate the economy and crowd out domestic investors from a position of command over the industrial sector need to be investigated. Once the need to analyze the interdependence of the two sectors, and the need to study them separately are recognized, still another important issue that needs to be addressed surfaces. That issue is the role of international trade. If the economy participates in 7 international exchange, in what way can it stand to gain by involving itself with the rest of the world? The con clusions arrived at by different economists vary consider ably in this respect. The question that needs to be investigated is: When would international trade facilitate the growth of the Lewisian economy and under what condi tions would the economy fail to harness the beneficial effects of foreign trade? Before proceeding to a more detailed discussion of the scope of the following essays, a few words need to be said about the methodology chosen. The four important issues are taken up one at a time in the following chap-1 ters. They are treated separately so as to highlight the distinct effects. A single model would pose problems in terms of distinguishing between various cross-effects and in the highlighting of the direct effects. The final chapter ties up the logical interconnections of the four different issues and shows how the contending perspectives on economic development relate to the Lewis framework. However, in writing the essays, an unifying connection is retained in terms of the basic characteristics of a Lewisian economy. The first essay concentrates on a fixed- 8 wage two-sector economy with the agricultural sector left sufficiently flexible to accommodate different types of organizational forms. The second essay also allows for different organizational structures in the agricultural sector. This is consistent with the fact that Lewis purposely left the structure of the non-modern agricultur al sector unspecified. The third essay allows the problem of foreign investment to be analyzed in a: situation where the supply of labor is perfectly elastic. Finally, the problem of international trade is analyzed in a fixed wage, labor surplus economy where its comparative advantage lies in the production and export of the more labor intensive good. 1.3 The Problem of the Terms of Trade A persistent criticism of the Lewis model has con cerned the internal terms of trade. If the capitalist sector depends on the subsistence sector for food, then as labor is withdrawn into the capitalist industries, the average productivity of those remaining in the agricultural sector would rise. It does not matter who reaps the benefits of the higher productivity; it could acdrue wholly or partly either to the landlords or to the pea sants. Similarly, accumulation in the capitalist sector would lead to higher profit income for the capitalists. Now, if any - not necessarily all - of the beneficiaries 9 of development have a positive marginal propensity to con sume food, the demand for food rises. Since agricultural output is constant, the price of food goes up. The terms of trade would move in favor of agriculture and the share of wages in the capitalist sector would rise. This would in turn lead to lower profits and hence to slower growth. Thus accumulation in the presence of surplus labor would push up wages, not just beyond a point, but from the very outset. All that is required for this problem to arise is that some people have a positive income elasticity of demand for food. An easy way out of this dilemma is to assume, albeit unrealistically, that everyone is satiated in their demand for food. An alternative is to assume a small open economy where the terms of trade are given, but that would be evading the issue. Only if the population were to grow at a very high rate would the consequences be ambiguous. The upward pressure on food prices will be countered by a swelling rural population that would reduce the average productivity of the subsistence sector. The structuralist and Marxist responses have been that this problem would inevitably arise in a private enterprise economy. Chakravarty (1977) noted that a possible solution to this problem was complete state control over the supply of wage goods, Taylor (1979) 10 points out that this was the outcome of the Soviet debate in the 1920s. Stalin found an effective way to tax away all additional income from the agricultural sector and he had the troops to do so. Mitra (1977) argues in a similar way that capitalist development is bound to taper off, the benefits being mopped up by the privileged few, while the bulk of the population is excluded from the development process. It is in this sense that the bonanza of rapid growth becomes translated into "luxury for the few seen as the fruit of misery for the many," [Taylor and Bacha (1976)]. Lewis was aware of the problems caused by an adverse movement in the terms of trade. In his 197 9 article he was quick to point out the implications of sectoral inter dependence. If a positive marginal propensity to consume food were to worsen the terms of trade for the manufactur ing sector, there would also be countervailing pressure on manufactured goods1 prices arising from the extra income flowing to the landlords, peasants and capitalists. The net effect on the product wage could therefore be in either direction. Lewis also mentioned that sectoral interdependence could make the supply of manufactured goods extremely elastic, hence making output demand-determined. The size of the harvest and farm incomes would naturally play a 11 large role in the determination of this demand. In this context, it would seem logical to rigorously analyze the supply constraint in agriculture and the effec tive demand for the manufactured good. To this end it would be especially interesting and important to work out the effects of changes in the marketed surplus and wages on the terms of trade and industrial employment. This is the task of the first essay. Results consistent with the Lewis expectations are obtained. The effects of these factors on the terms of trade and employment turn out to be ambiguous, the actual realized sign of the effect being critically dependent on the relative magnitudes of the marginal propensity to consume of the various groups. In fact, the possibility of an unstable situation also emerges, with the economy experiencing cyclical fluctua tions in prices and employment. The results demonstrate that, contrary to what has been suggested by some of the critics [Guha (1969)], there is nothing inevitable about the breakdown of the growth process. The inelasticity of agricultural supply and the problem of effective demand in manufacturing are the cru cial problems of a Lewisian economy. If the supply of food is a constraint, then technical change and innovation in the agricultural sector becomes of paramount importance. Similarly, the problem of demand can be tackled by aug menting domestic investment through foreign aid and 12 investment- These are not new prescriptions, but just how they fit into the scheme of Lewis needs to be investigated. Finally, the possibility that international trade can contribute to the development process needs to be investigated. If domestic accumulation bears heavily in terms of its costs, why can't the economy bow to the dic tates of comparative advantage and international specializa tion? Manufactured goods could be imported and the accumulation process facilitated. These are precisely the questions posed in the last three essays. However, it is necessary first to address the issues in the light of criticisms made. 1.4 Technological Change in Agriculture The importance of technological change in the agri cultural sector of the Lewisian economy has already been noted. Lewis purposely left the structure of this sector unspecified, noting that it would vary from country to country. Indeed, a wide variety of institutions can be found in the agricultural sectors of less developed countries. Few of Lewis' critics have agreed on the "climate" and institutional framework that most favors technological change. Some [e.g., Schultz (1964)] have noted that the availability of new technology is the 13 crucial question. Farmers would respond to productivity increasing inputs, provided they are cheap and readily available. Others, like Bhaduri (1973) have noted that the structure of agrarian relations can act as a fetter to new investment in technology. The only way to break the "fetter" would be a thorough-going revolution in property rights and institutions. Once again, the polar view of the critics is noted. In the Lewis framework, either the availability of new technology is sufficient to accelerate growth, or the rigidities of the agrarian structure prevent it. These extreme cases seem to emerge as distinct pos sibilities, once costs other than the directly observable ones are admitted. Drawing on the literature on trans actions costs, the second essay demonstrates the fact that new technology invariably calls for new contracts between agricultural agents, new problems of monitoring and enforcement; in short, a new institutional set-up may emerge. If these costs are too large, a new technology may be rejected even though it passes the criterion of higher profitability on a direct cost-benefit computation. New technology may or may not be absorbed, and each case has to be studied on the basis of its specificity. The problems of technology and institutions are closely con-- nected. There is no guarantee that technological change 14 will always be forthcoming. 1.5 The Issue of Foreign Investment In the Lewisian economy, capital accumulation in the industrial sector crucially depends on how much of the profit income is saved, and how much of the savings actual ly gets translated into investment. A closely related issue is the need to look for productivity increasing technology, to overcome the bottlenecks to capital forma tion. It was noted earlier that a possible way of increasing aggregate demand and benefitting from a better technology is to allow foreign investment. Here, also, the literature has diverged to extremes. Foreign aid is seen either as a most useful vehicle for rapid development, or as an instrument of domination by the developed coun tries whose investment crowds out domestic investment. Indeed, it is often viewed as a tool by which developed economies can gain complete control over the host economy. The charge of domination is closely linked to the objective of self-reliance in economic growth. The third essay tries to clarify this issue. It is noted that foreign investment can only be realized at the request of the host country. This implies that contractual agreements have to be made and transaction costs incurred. In fact, the very act of investment itself may require new 15 contracts implying the significance of transaction costsl An important result that is obtained proves that foreign investment embodying a superior technology may completely dominate the economy. Whatever be the equilibrium magnitude of foreign investment, its contribution to growth is unambiguous. A reduction of transaction costs faced by domestic investors and an increase in those faced by foreign investors can contribute to growth and also reduce foreign control. There is no trade-off between self- reliance and faster growth. Within this general framework, it is the nature and magnitude of transaction costs that determine whether foreign investment becomes the all- powerful driving force in the economy, or whether it acts as a mere catalyst. 1.6 The Outcome of International Trade Economists have observed that international trade is a non-zero sum game and all participants stand to gain. An economy trying to develop can exploit these potential benefits by opening its borders to the flow of goods and services from the rest of the world. Lewis was aware of this, but he noted that reliance on trade alone may not be sufficient and a conscious effort must be made to build up a domestic industrial structure. Authors of a more radical persuasion have argued that international trade contributes to the thwarting of the growth potential of the less developed countries. Trade is a zero-sum game. Development is implicitly equated to the growth of a manufacturing sector. Hence, the absence, or the gradual demise of the manufacturing sector is the sign of under development. However, the standard Ricardian and Heckscher-Ohlin models of trade focus on the gains from trade in terms of the improved availability of goods and services. It does not matter where the goods are produced. The location of the manufacturing sector and the international division of labor is decided by the law of comparative advantage. Therefore, the lack of a domestic industrial structure is neither necessary nor sufficient for underdevelopment to occur. Both points of view require some clarification. In a dynamic context, to illustrate that trade can ruin an economy, the proponents of uneven development must prove that trade jeopardizes the entire production structure of the economy. On the other hand, proponents of the Heckscher-Ohlin view must realize that the existence of the gains from trade is not enough; the critical question is what is done with these gains. The gains can be frittered away on consumption of non-essential luxury goods, or they can be utilized to enhance the productive capacity of the economy. The implications for long-run growth are quite 17 different in the two instances. The final essay attempts to highlight these issues. It is shown that the gains from trade are present in a neutral fashion. However, unequal development is a pos sibility when the gains are not harnessed to the structur ing of future productive capacity. Seen in the light of sectoral interdependence, unequal development has a harsher implication than noted by the radical authors. It can undermine all production activities of the economy. Lewis's (198 0) suggestion about the importance of building up a domestic industrial structure while simultaneously enjoying the benefits of free trade, takes on new signifi cance . 1.7 Concluding Remarks To sum up this rather lengthy introduction, the essays attempt to accommodate the critics of Lewis within the framework of the Lewis "vision". The problems and prospects for development are not unique and cannot be capsuled into an easy recipe. Each economy may have its own distinct set of problems. It calls for, implicitly, more empirical studies of concrete situations, before any meaningful policies can be prescribed. A paradigm, if it can be developed at all, must be able to encompass alternative possibilities. The rich variety of history 18 bears evidence to this idea. The Lewis framework could be a starting point. Leeson (1979) puts it aptly: Possibly if those who sought to annihilate it (the Lewis model), and one another, in the course of the evolution of development economics had instead built on it the discipline would not now, twenty-five years on, be so characterized by the mutual incomprehension by each school of thought of the work of the other. (p. 209) 19 CHAPTER 2 THE LEWIS MODEL WITH FOOD SUPPLY AND AGGREGATE DEMAND CONSTRAINTS 2.I Introduction Some recent authors like Bharadwaj (197 9) have noted that the stylized version of the Lewis growth process is apparently unsatisfactory in explaining the experience of typical less developed economies. Recent empirical work on India has revealed that there is no -tendency for the predictions of the dual economy model to occur. Over the past three decades employment in the modern sector has increased very slowly even though the real wage-rate has remained more or less constant. Yet the terms of trade moved increasingly in favor of agriculture. This has been observed even during periods of high agricultural output and technological change. Investment activity in the modern sector has been sluggish, yet rates of profit have not shown a sharp decline. In short, growth has not been uniform and sustained. While the employment paradox could be explained by the movement of the terms of trade and possible labor- saving innovations, there are two major problems that remain. First, if the marketed surplus of food is crucial, and mere increases in total output do not guarantee an 20 easing of the food constraint, then the agricultural sector and its structure must be explicitly analyzed and not be treated simply as a part of the "black-box" called the traditional sector. Secondly, the sluggish pace of investment in the face of fairly constant rates of profit indicates that the assumption of all savings being auto matically invested, must be questioned. If savings and investment are not matched in an ex ante sense, then effective demand constraints must be explicitly analyzed. Gersovitz (1974) provides some explanation as to why ex ante savings may not be equal to ex ante investment in a developing economy. While Gersovitz is able to obtain an endogenous determination of the wage-differential between the two sectors, he avoids the need to analyze the impact of effective demand on the terms of trade by assuming that the economy is a small open one, with given terms of trade. Thus, any meaningful extension of the Lewis framework would call for an explicit analysis of the agricultural sector, endogenous determination of the terms of trade and the incorporation of effective demand constraints. The food constraint together with the problem of demand could have an important effect on the terms of trade. This essay develops a two-sector model along the above-mentioned lines. It is assumed in the Keynes-Clower 21 tradition that agents are quantity-constrained and, hence, their demands are functions of income rather than price. The model follows the tradition of Kalecki (1976), Taylor (1979),and Rakshit'(1982), in the sense that, while deficiency of demand is the main problem in the modern sector, it is supply that is the chief constraint in the agricultural sector. This implies that an increase in demand would raise industrial output and push up the price of food. In short, the adjustment mechanisms in the two markets are different. In Section 2 the model is outlined and the macro equilibrium of the system analyzed. Section 3 deals with the comparative statics results. Section 4 notes the significant results and points out the possibilities of future research. 2.2 The Model In constructing the model, the following assumptions are made. (i) The economy consists, of two sectors, agriculture producing food and the industrial sector producing a manufactured good. (ii) There are five classes: landlords, peasants, capitalists, workers and merchants. 22 (iii) Merchants only trade, others consume both goods. (iv) The industrial wage rate is fixed and given in terms of food. (v) The economy is closed and the terms of trade, P, P! is the food price of the industrial good, i.e. P = = 5— . F (vi) There is imperfect information in the sense that trading can take place at false prices. In other words, agents are quantity-constrained and, since there is no auctioneer, recontracting is not possible, except in the case of the merchants. (vii) The merchants are not quantity constrained, and it is their demand for food which is a function of price that clears the food market. (viii) Given assumptions (vi) and (vii), the terms of trade between industry and agriculture change only because the price of food changes. In the industrial sector supply adjusts to demand. The Agricultural Sector The landlords and peasants jointly organize produc-'. tion, given the technology and a contractual rule for determining total output A and its division. Let the landlords' share be aA and the peasants' share be (l-a)A, where a is a fraction 0 < a < 1. The landlords consume a 23 part of their income and with the remainder demand-the industrial good, which is denoted by R(aA). Similarly, the peasants' demand for the industrial good is given by T((l-a)A). Hence, the total marketed surplus is given by R(aA) + T{(l-a)A}. It should be noted that the marketed surplus is a fixed quantity that depends on the total output of food and its distribution within the agricultural sector. This implies that the price elasticity of the marketed surplus is zero. The way the agricultural sector is modelled is simplistic but it helps to focus on the problem of the distribution of income within agriculture, emanating from a variety of organizational forms. The Trading Activity of Merchants The merchants are only interested in the trading of food. Their consumption demands and incomes are ignored. Their demand for food is for speculative purposes and is price dependent. In an dynamic setting, this would be interpreted as hoarding and dishoarding, but for the present their one period demand is considered. Since the food market is usually tight in the developing economies, prices adjust faster than quantities, but not fast enough to eliminate the possibility of speculative gains alto gether, Indeed, this activity may help to stabilize prices. Alternatively, since this activity requires 24 superior information about prices and-demand^ it couIS-also be thought of as state trading in food. In that case, however, the purpose would have to be reinterpreted as 'one of building buffer stocks. This demand is denoted by: IF (P), l£(P) > 0. The Industrial Sector The capitalists and workers organize production in the industrial sector, given the production function N = NCLj), where N is the industrial good and Lj the industrial employment. The wage-rate Wj is fixed in terms of food. Aggre gate supply, in the notional sense, is determined by the dN dL. profit-maximizing condition PN' = WT, where N' = > 0, I and N" < 0. The profits, and hence the income of the ^1^1 capitalists, is given by ( N - — =— ) . The capitalists' consumption and investment demand for p w il i the industrial good is given by C ( N - — =— ) and WILI I( N - — =— ), respectively. Their demand for food is WILI given by CF ( N - — F— ). The workers' income is W ^ . Their demand for the industrial good is given by WfW-j-Lj). The residue of their income is spent on food, given by (WjL^ - WCW-j-L^)}. Note, this is equivalent to saying that they spend constant 25 fractions of their income on food and the industrial good, The aggregate demand, AD, for the industrial good is given by: (2.1) AD = R(gA) + + C ( N ) W L W(W L ) + I ( N - ~r ) + =----. Because of quantity constraints, output of the indus trial good is demand-determined, so: (2.2) AD = N The notional aggregate supply of the industrial good is: (2.3) PN' = Wj The food market balance is given by equating the supply to the demand: W L (2.4) R(aA) + T{ (1—a) A} = Ip (P) + PCp ( N ) + WjLj. - W(W Lj) Putting a = a and A = A , and substituting (2.2) into (2.1), the following is obtained: 26 (2.i>) R(ot A) +^T{ (1 a) A} + c( N _ ) W L W(W L ) + I ( N = N Now, using (2.1') and (2.4), we can solve for P and N. Because of the Keynesian nature of the model, the P solved from (2.1') and (2.4) will not in general satisfy (2.3). In general, PN' > Wj or w 1 - Pip- > °- It is assumed that a solution to (2.11) and (2.4) exists. Differentiating (2.1') yields, W [ (1-1'-C') + (I'+C'-W')] dP dN ^ iy [R+T+W-WtL (I' + C')] P^ The denominator is > 0 from (2.4). The numerator is > 0 if (I' + C') < 1 and (I* + C') > W'. Hence, (I' + C') < 1 and (I' + C ) >_ W' are sufficient to ensure a negative slope of AD in the (P,N) plane. It is assumed that the 27 sufficient conditions hold. Differentiating (2,4) yields: dP dN W W {CF ( P " > + ST (1“W,)} r OOQ , _ , i p i p t \ ' F F F P dP Cp 0 and W 1 < 1 are sufficient to make negatxve. Food Since W' <. 1 has already been assumed from above, because W £ I' + C' £ 1, the only additional assumption required dP for assuring ^ negative is C' > 0. Food From (2.3) it is noted that given Wj, the higher is P, the lower is N' and hence the higher is N. Therefore, the AS (aggregate supply) must be upward rising in the (P,N) plane, The adjustment mechanism obtained from assumption (viii) is given by: = F (AD - N) F (0) =0, F' > 0 dP F F = G(D - Sr ) G(0) = 0, G' <,0 where DF and SF are the demand for food and the supply of the marketed surplus, respectively. 28 From the above two differential equations, the stability conditions obtained are: (a) F' [R + T + W - WjLjd' + C + — (1 - W' ) } > 0. N' <b) P 1 2 W W [R+T + W-WjL 1 (I' + C' ) ] {C£(P - ^ ^ (1-W')> > 0 . Condition (a) holds if I' + C' <_ 1, C' > 0 and W' 1. — r — — This follows from the assumptions already made regarding the slopes of the aggregate demand and food constraints. Condition (b) holds if the absolute value of the slope of the aggregate demand constraint exceeds that of the food constraint. Given the nature of the slopes of (2.1)', (2.3) and (2.4), and the implication of the stability conditions (b), the macro-equilibrium of the system can be studied. The three possible cases are each studied in terms of a simple diagram. The AS curve can lie either to the left or right of the intersection of AD and food, or it may pass through the intersection of the two. 29 Case (i) ; AS II Food III AD Figure 2.1 Excess Supply At III AD = AS, but DF < SF. So P rises to clear the food market. At II, DF = SF, but AD < N, so N falls. At F F I, D = S and AD = N. N will not increase because the extra output cannot be sold due to insufficient demand. I is an under-full employment equilibrium, with P and N not tending to change. This may not imply open unemploy ment if those who could not obtain industrial employment became peasants. The situation basically implies that the ______ 30 full potential for industrial expansion is not being realized because of a lack of demand. Case (ii) P AS III II Food AD Figure 2 .2 Excess Demand At I, AS is not satisfied and the price-output combination is not feasible. At II, AD > N, but N = AS, IT F hence P will rise. At III, AD = AS, but D > S , so P will fall. 31 The net result is that P will either fluctuate between II and III or stabilize somewhere between II and III as a quasi-equilibrium, with PT and P , . , both rising, but X r P remaining unchanged. The final outcome depends on the speeds of price adjustments and the magnitudes of the responses in the two sectors of the economy. Case (iii) P „ AS p* Food AD 0 N* Figure 2.3 Full Equilibrium In this case, all three constraints are satisfied, implying that all plans are realized. 32 2.3 Comparative Statics Differentiating (2.1') and (2.4) and solving them simultaneously, the comparative statics results for changes in A, a and can be obtained. Since the algebra is elaborate, the derivations are provided separately in Appendix 2.1. Here, only the results are discussed. Change in A, The Agricultural Output The economic reasoning is clear. An increase in A which can be thought to be brought about by an improvement in agricultural technology, increases both the level of aggregate demand and eases the food constraint. In terms of the previous diagram (Figure 2.1), both AD and the food constraint shift upward. The net result depends on the magnitude of the shifts. It would be expected that an easing of the food constraint would increase industrial output and move the terms of trade in favor of industry. < 0 as PI' + WtLt + W_L_(C1 + I’ + C) dA F I I I I F < R + T + W - PC. F and W i ’ - c - c p ( p - ^ ) r 0 33 Since nothing has been assumed that guarantees the fulfiTl” ment of the sufficient conditions, a situation may arise where an increase in agricultural output could move the terms of trade in favor of agriculture, as observed by Bharadwaj. This would happen if the price change required to clear the food market is small and the response of ag gregate demand is sufficiently high.. Change in a, The Landlord1s Share of the Output clN^ clP The signs of ^ and ^ are also ambiguous. > 0 if R' > T' and da------------------------- --- PI' + WtLt (C' + I' +C')>R+T+W- PC„. F I I F F < 0 if R' > T' and da , ------------ PI' + WtLt(C' + I' + C) < R + T + W - PC_, F I I F F or, if R' < T' and PI' + WtLt (C' + I' + C) > R + T + W - PC„ F I I F F dP ~ > 0 if R' > T' and da --- W ( P “ jgT" ) (1 ~ I' ~ C' - Cp) > 0. dP < 0 if R' > T' and da — - W ( P - ^ ) (1 - I* - C' - cp < o. 34 or if R' < : T ' and W ( P - ) (1 - I' - C' - c p >0. A redistribution of agricultural income, as for example, brought about by a change in the type of contract, may or may not increase the marketed surplus of food, and hence the agricultural sector's demand for the industrial good. Depending on which direction P has to move to clear the food market, the net effect on aggregate demand is determined. The final effect on (P,N) depends on the relative magnitudes of these responses. Change in W^, The Industrial Wage-Rate The signs of and are ambiguous. A sufficient ^ dWj dWj - condition for < 0 is (1-W") < PC^,. If this holds, dP ^ then g— > 0 always. An increase in the wage-rate reduces employment, but moves the terms of trade in favor of industry. As before, the marginal propensities to spend are of crucial significance. It is in this sense that the distribution of income within industry and within agricul ture is so important in determining the behavior of the economy. 35 2.4 Concluding Remarks It may be stated that the model described above is a more general version of the stylized Lewis model. It highlights the possibilities of a wide variety of results which Lewis emphasized in his 1979 article. The develop ment of the modern sector may not be a smooth and balanced experience. Indeed, the economy can experience cyclical fluctuations in P and N as proved in Appendix 2.2. It is crucial to understand how different markets adjust and how different social classes respond to a change in their incomes. The Lewis model highlights the possibilities of long term growth in a labor-surplus economy. The model de scribed in this essay highlights the short-term or medium-term problems that may arise in the process of a Lewis-type development in the economy. Lewis (19 79) himself was aware of the importance of distinguishing between short and long-term effects. Hence, the static framework of the model is not totally inconsistent with the long-run dynamics of Lewis. Within this short or medium-run setting, not much was said about the way agricultural output, A, was determined. One possible way is to endogenize A, and explicitly incorporate the role different labor-land contracts play in determining it. One way of studying the 36 role of contracts and different organizational forms would be to build a sub-model for solving A. Then, with specific functional forms of the demand functions, the system could be solved numerically. The nature and magni tude of the divergences of the solved (P,N) from the full- employment equilibrium price-output configuration, (P*,N*) could be examined. Two other things remain to be said. One is regarding the allocation of labor, migration and unemployment that was implicit in the model. The other comment pertains to the multiplicity of results. In Case (i) of the model there was potential unemploy ment. Suppose there is surplus labor in agriculture. If there is migration to the industrial sector, food output will not be affected. Industrial employment will be con stant and there will be open urban unemployment, in the fashion of the Harris-Todaro model. On the other hand, if there is not surplus labor, but migration still occurs in response to higher expected incomes, two things might happen depending on whether g|^ <0. If due to a fall in A, J Vf N also falls, i.e. g-^- > 0, unemployment will be larger than the volume of migration. In other words, migration may results in some of the existing workers losing their jobs. If g|- < 0, some new jobs will be created in the industrial sector and unemployment, if it exists at all, 37 will be less than the volume of migration. In this case, migration on the basis of expected incomes may actually result in the creation of new jobs in the industrial sector. Finally, a few words need to be said about the different possible outcomes of the model. A simpler model that fails to explicitly recognize the demand constraints could end up with the result that the terms of trade would always move against the industrial sector from the very onset of the accumulation process . Lewis had observed that the demand side would have a countervailing effect on the terms of trade. This would be particularly important in the short and medium-run time horizons. The final effect on the terms of trade would be determined by the interaction of both demand and supply factors. The multiplicity of results occur at two levels. The first is in terms of identifying the nature of the tempo rary equilibrium obtained, whether it is a case of excess supply, excess demand or full-employment, These spell out the theoretical possibilities. For a Lewisian economy, marked by open or disguised unemployment, the case of excess supply is perhaps the most relevant one. It should be noted that this case is not inconsistent with the observed fact that less developed countries often experience high rates of inflation along with high unemployment rates. In the case of excess supply, the relative price could increase if the underlying parameters 38 of the aggregate demand constraint and the food constraint changed. The second source of multiple results is in terms of the signs of the comparative statics results. In general, the signs are ambiguous and sufficient conditions had to be imposed. Given this source of multiplicity, it is use ful to make a comment on the plausible results that would occur.in a Lewisian economy. Demonstration effects and a time-preference pattern favoring current consumption would imply high marginal propensities to consume. High marginal propensities to consume would tend to yield results that were in apparent contradiction to a simple version of the Lewis model. It could produce results like the one observed by Bharadwaj (1979) , where the terms of trade moved in favor of agriculture even during years of good harvests and higher agricultural output. Sufficiently high marginal propensities to consume could also make the equilibrium locally unstable, and lead to fluctuations in the relative price and employment. 39 APPENDIX 2.1 m , dP 1. To find jtj AD Totally differentiating (2.1)' and setting da = dA dN dWj = 0 and using dLp = j^r- or W W L dN = - -% dP + I'dN - I'pjfr dN + I' dP W W L + C'dN - C* ==£- dN + C* dP - dP PN' p 2 p2 W + W -fr dN - 4r dP w w w dN[ 1 - I* + I' - C* + C' ^ - W ^ ] dP [ -R - T - W + I’WpLp + C'WtLt ] I I or dP = dN W [(1 - I' - C’) + ^ (I' + C' - W')] =j [R + T + W - WpLj (I' + C ) ] 40 the numerator is the denominator positive, is positive dP Ce' dN dP To fxnd JTT dN Food Totally differentiating (2,4) and setting da dA dW dN + PC' dN dP + C„dP or dN dP or dP dN 41 p - N' W ) + ^ (1 - W ’)} I* + C_ + C' F F F WILI The denominator is positive if Cp > 0. 1 - W" _> 0 and, ap generally, P > • So' gpj < 0 if cp - 0 # 3. Stability Conditions. dN d = F(AD - N) F(0) = 0, F' > 0 S- = G(DF - SF) G (0) =0, G' < 0 dt The characteristic equation is 1 WI -F'iA- [R+T+W-W tL (I'+C )] }-A F' [l-d'+C1 ) (1- p^r) p wT - W 1 — 1 PN' J WtLt WI WI G'(I^+Cf+C^ ) G- [^(^-P)- ^(1-W')]-A] 0 42 w => (a) F* {-^-[R+T+W-WjLj (I'+C' ) ] } - G'[C£(P - ) W + (1 - W')] > 0 W L W (b) (Ip+Cp+Cp )[(1-I'-C') + pjp- (I'+C'-W)] w w [R+T+W-WjLj (11 +C ' ) ] [Cp (P — (1-W')]> 0. or W W W [(l-I'-C) + p^r (I'+C'-W*)] C£{P-jjf) + ^ !(1-WV) 1 - W L ~ [R+T+W-WtLt (I'+C1 ) ] , I I x x F F F P Slope of AD I > | Slope of FC 4. Comparative Statics. Let dP dN AD P. 3 43 W [(i - i* - o + p— d' + c - wv)] ■ — [R + T + W - WjLp (I' + c')] Let dP dN Food _ X. 6 Change in A. {C^(P Wp N' ) + Wp N' (1 - W' ) } W L ( ' I' + c + c1 —-—— ) F F F P ’ dA = -gdP = pdN ... (i) SdP = -ydN + qdA (iif where q = aR* + (1 - a)T'. (i) and (ii) are obtained from differentiation of (2.1)1 and (2.4). Solving (i) and (ii) simultaneously, we can obtain and Note p6 Y$ dA dA ‘ A is the stability condition (b) Hence, A > 0. From (i) 2. dA - pdN = BdP or dP = dA- 3- dN. From (ii) dP = -X dN + 9- dA or AdN = 2 (5 - pg)dA, ® a - pbi 6 6 P dN P ( or dA A 44 ^ < 0 as 6 < Pg, since q and P are both positive. dN > - aa < 0 as ^1^1 > 1 XF + CF + CF ~P < p (R + T + W) WILI (I' + C') Rearranging dN > 0 as PI' + WtLt(C' + I’ + C’) dA < F 1 1 F E + T + W - PC„ F From (i) dN = — dA - — dP Pp p From (ii) dN = — dA - — dP y y r - . 1 J Z dP Solving for 3. (g _ p D ) dp p dA „ dP > r \ & ‘ k n dP > A So, < 0 as 6 > Pp. < 0 as ' dA K dA W W Cp (P - ^ ) + ^ (1 - W) i P(1 - I* - C) 45 AV w + — (I• + c - w') w So, || < 0 as (1 - I' - c' - c p ( P - ) < 0 Change in a. pdN = -BdP + | da (iii) SdP = -ydN + tda (iv) where t = (R' - T')A From (iii) dP = da - dN From (iv) dP = - ^ dN + da dN I (6 - Pe) da > 0 if t > 0 and S > P3 da --- < o if t < 0 and 6 > P3 da or t > 0 and 6 < P3 46 A sufficient condition for < 5 > Pg is WILI 1 ' I X S + Cp + Cp > p- [R+T+W-WjLjd'+C')] or Pip + WjLj (CJ, + I' + C ' ) > R+T + W- PCp , Solving for dP. I (pp - Y) da A > 0 if t > 0 and Pp > y da ^ < 0 if t > 0 and Pp < y da or t < 0 and Pp > y A sufficient condition for Pp > y is W ( P - jjf ) ( 1 ” I' - C' - Cp ) > 0 47 Change in W^. ^1 pdN = -BdP + ~ X dWj (v) -ydN = 6dP + LjZ d Wz (vi) where X = (W1 -I' - C') < 0 and Z = (1 - W' - PC£). Solving for dWz - LI dP _ T" (ppz + yX) dW^T ~ ' A (Y, P) > 0 X < 0 (LjfP) > 0 dP So, a sufficient condition for -grr- > 0 is Z <0, i.e. ' dWT dP 1 - W' < PC'. Note if Z > 0, - sty- is of uncertain sign, r uWj Solving for ^ , we obtain LJ.X6 + LIZP3 dN ^ P dWz A (A, 5, g, P, Lz) >0, X < 0. So g§- < 0 if Z < 0, i.e., 1 - W' < PC£. Note if Z > 0, is of uncertain sign. 48 APPENDIX 2.2 The dynamic system; F (0) = 0 F' > 0 G (0) = 0 G' < 0 was analy2ed to obtain the stability conditions of the macroequilibrium, if it existed. Stability conditions (a) and (b) were equivalent to having the trace and determinant of the Jacobian matrix negative and positive, respectively. It is noted that if I' + C* is sufficiently large, so that the trace is positive, (i.e. condition (a) is < 0) without changing the sign of the determinant, then a locally unstable equilibrium is obtained. The global dynamics implied by the locally unstable equilibrium will lead to cyclical fluctuations. This can be proved with the aid of the Poincare-Bendixson theorem and assuming that all functions in the system are continuously differentiable and monotonic, and that the domain of the above dynamical 2 system can be chosen to be a closed compact region of 1R+, diffeomorphic to a disc. The Poincare-Bendixson theorem dN dt = F (AD - N) dP dt G (D SF) 49 states that a non-empty compact limit set of a continuously differentiable dynamic system in the plane, that contains no equilibrium point, is a closed orbit. Since the equilibrium is locally unstable, it is not a limit point and hence not an element in the set of limit points. There is no reason why such an equilibrium should be ruled out. The marginal propensity to spend being greater than unity is plausible in the short-run. Equilibrium models that focus only on comparative statics or local dynamic analysis make little sense if the equilibrium is unstable. However, it is important to investigate the nature of the underlying equilibrium. Since the model describes a temporary equilibrium, it is plausible that the underlying parameters change in the long-run. Learning by the private sector may remove these oscillatory tendencies, the cyclical transition stage of adaptation cannot be ruled out. It provides an explanation as to why a less developed country may exhibit fluctuations in P and N within limits, without "taking-off" into self-sustained industrial growth. 50 CHAPTER 3 TECHNOLOGICAL CHANGE IN BACKWARD AGRICULTURE 3.1 Introduction It was seen in the previous essay that the size of the marketed surplus from the agricultural sector was important in determining the level of industrial output and employ ment. In this context, new technology of agricultural production could contribute to the increase of the marketed surplus. Total agricultural production must be increased if the growing industrial labor-force is to be fed adequately and cheaply. This was clear in the case where an increase in the agricultural output contributed to higher industrial employment. This is the most likely outcome when marginal propensities to consume are high. In the more unlikely case where a rise in agricultural output reduces indus trial employment, technological change and higher output would have the welfare effect of raising the incomes of those engaged in the agriculture sector. However, the mere availability of a superior technology is not suffi cient to guarantee its adoption' by agricultural agents. It is argued here that the nature of the existing contracts and the magnitude of transaction costs are critical in 51 determining whether a new technology is used or not. Before presenting the analysis, a brief review of the literature is necessary. The macroeconomic literature on development has almost always treated the agricultural sector as being homogeneous in structure. In this respect, the impact of the Lewis model was, perhaps, significant. Writers dealt with the problems of stagnation, primitive technology, unemployment and underemployment. Analysis was very seldom focused on the microeconomic structure of agriculture, particularly during the decades of 1950 and 1960. Any restructuring of the backward sector did not matter in the Lewis model, as long as it remained a reservoir of surplus labor. The few micro-studies that were done [e.g. Schultz (1964)] , concluded that farmers were poor but efficient, and the problem of the backward agricultural sector could be largely solved by the cheap and ready availability of a new technology. During the late 1960s, when the new technology embodied in the Green Revolution, was experimented with; it was observed that the superior technology was not uniformly absorbed. The consequence was the abundance of more detailed microeconomic studies on less developed agriculture, during the last decade. Research was directed to the analysis of the different types of contracts that 52 were observed. A number of issues were raised. Were all the different contracts that were observed efficient? If some of them were inefficient as optimal allocations of scarce resources, why were they observed to coexist with the more efficient ones? What were the distinguishing characteristics of the markets? What was the significance, if any, of non-market forces? A full review is not attempted here. Only certain hithe-rto neglected elements, that have now become integral parts of the general theory of contractual choice, are discussed. The first is the concept of transaction costs. Every transaction involves a contract; and every contract involves negotiation, enforcement and supervision costs that must be incurred by at least one of the contracting parties. An explicit recognition of these costs must be made in analyzing the actual allocation of resources, implied by different contracts. Secondly, economists have increasingly used the concept of entrepreneurial ability, and its difference across individuals, in explaining the coexistence of different types of contracts. A closely related, though distinct, concept is risk-sharing. Individuals vary in their ability to bear risk. In the face of uncertainty, they may not be able to bear all of it. Some may wish to share it in the most efficient manner. Still others may 53 choose not to bear any risk at all. Finally, since every contract is a transaction and involves at least two parties, information about each other becomes crucial. Will the individual keep to the terms of the contract or will he shirk? If these qualities are not directly observable, but have only imperfect surrogates, then the acquiring of maximum information becomes extremely important. Given the heterogeneous distribution of entrepre neurial abilities, and information, what distinguishes one contract from another is the identity of the residual claimant. For example, in a fixed wage contract the land lord claims the residual; whereas in a fixed rent contract, the tenant claims the residual. In a sharecropping arrangement, they share it. While contributions to the theory of contractual choice have been substantial, relatively less attention has been paid to the dynamic question of technological change. Allocative efficiency questions are analyzed in the context of a given technology, but it is equally impor tant to study the response different contractual arrange ments have to the availability of a new technology. In this context Bhaduri (197 3) made an important contribution. Section 3.2 uses the concept of transaction costs to pro vide an explanation as to why a particular technology may not be absorbed, whatever be the nature of the contracts 54 existing in backward agriculture. Section 3.3 shows spe cifically for sharecropping arrangements, the possibility of a new technology not being adopted. The final section has some concluding remarks. 3.2 The Bhaduri-Newbery Debate This section develops the arguments of Bhaduri and Newbery (1977) about the possibilities of technological change and innovation in a backward agricultural sector, marked essentially by the dominance of a landlord class with monopoly-monopsony power. An alternative framework is suggested in which the issues raised can be studied. Bhaduri pointed out that under semi-feudal agriculture the big landlord was also the moneylender, and used the instrument of usury to exploit tenants and workers. The demand for consumption loans and the timing of the loan- contract gave, as it were, an additional instrument for accumulation to the landlord. Any technological change would have the effect of making the tenants better-off and hence reducing the demand for loans. The resultant loss of usury income, Bhaduri pointed out, would act as a dis incentive for innovation. It may be noted, however, that this framework does not imply the absence of innovations. It merely indicates that only those innovations would be introduced where the loss in usury income would be more than compensated . for by the rise in profit income. Newbery pointed out that while non-economic reasons 55 for not innovating may exist, Bhaduri's economic argument was implausible. If the landlord could control the sharing of output (wages or the rental rate), then by using his monopolistic power he could force the tenant or worker to a subsistence level. If such was the case, then the land lord could also extract all the potential benefits accruing to the tenant-worker as a result of innovation. The essen tial argument of Newbery was that a new technology may im ply the need to devise new contracts, and hence may not be actually implemented, but his emphasis was on non-economic reasons. The upshot of the debate points to the fact that con tracts in agriculture are complex. For example, consump-' tion loans may be viewed not as an exploitative mechanism, but as an insurance or side-payment in the presence of un certainty. Furthermore, contracts are technology specific, hence a new technology calls for a reopening of contracts and the incurring of substantial transaction costs. A useful way of looking at contract formation is by viewing it as the outcome of a game with side-payments and differential information. Some work has been done in this area [Bell and Zusman (1976), Kotwal (1981)] but it has not been used to analyze technological change in the presence of transaction costs. In suggesting a game-theoretic framework, I begin with a simple competitive economy to define a core allocation. Then the conditions are reinterpreted to explain the outcome as one thrown up by a backward agricultural sector. It is assumed that there are n landlords and m tenants (workers). Let F(L) be the production function, where L is labor. Consider, given technology, the maximum output any coalition of |l | workers and |K| landlords can produce, by breaking away from the existing arrangement. Competition => F'(L) = p- (the real wage) Equilibrium => D L SL (demand for labor equal to the supply of labor) or nL = m or L m n Let the tenant's income be X. Then X = F' ( - ). v n Let the landlord's income be Y. Then - F’( - ). n n Maximum Coalition Output: V(L,K) = Max F(Lj) j<rK Subject to E ' L . = |L j eK 3 57 This implies L^ , where denotes absolute figures. So, V(L,K) = IK1 F ( --- ). The arrangement (X,Y) is 1*1 a core allocation if it is feasible and stable, that is, if there is no coalition that is better-off under any alternative allocation. So, (X,Y) is a core iff: Condition (3.1) implies feasibility. The total output produced is exhausted as payments to landlords and workers. Condition (3.2) is the stability condition. The left-hand side of the inequality shows the total income earned by some coalition (|L|, |K]) under the existing arrangement. The right-hand side shows the maximum output they could have produced (and earned as income) if the coalition (|LI,IKI) broke away from the existing arrangement. If their current income is at least as great as what they could have earned under an alternative arrangement, there is no incentive for any coalition to break away. (3.1) mX + nY = nF( - ) n and (3.2) | L t X + | K | Y > | K ( F ( -- Ik V (L,K) 58 Now, (X,Y) can be interpreted as being feasible and obtained not by competition, but as the outcome of a principal-agent game. So, stability implies that for all coalitions (|l |,|k |), condition (3.2) holds. However, (3.2) implies that alternatives are considered only under the same production technology, F. Now consider a new technology G: G > F V L, but: |L| | L I X + |k |y I >_ [ K I G ( ) , |K| then the original arrangement is stable. In this simple framework it can be seen that not any technological possibility will be grabbed by agents. A change to a potentially new and superior technology calls for the reopening of contracts and hence transaction costs have to be explicitly recognized. Assume that the transaction costs implied by the new technology is a function T(G), determined by the new technology, G. Suppose that the new technology has the potential for destabilizing the original arrangement, i.e. | l| |l ]x + |k |y « , |k |g ( --- ). I K I However, once transaction costs are explicitly recognized, the original arrangements may be stable, i.e. 59 ( L | X + | K | Y ^ |K|G( --- ) - T(G). |K| In this situation the new technology may not be accepted because of high transaction costs. The conjecture here is that the more fundamentally different G is from F, the higher will be the perceived transaction costs, and more the possibility of the new technology being rejected. This will not arise from merely the landlords' point of view, but as jointly perceived by landlords and tenants as partners in a contract. However, the result obtained crucially depends on the necessity of reopening of contracts, when agents are faced with a new technology. The next section takes up the possibility in the specific case of sharecropping. Sharecropping is accepted as an observed fact, and an explanation for its existence is not offered. 3.3 New Technology under Sharecropping Contracts are differentiated on the basis of who becomes the residual claimant. The residual claimant is the terminal input buyer and obtains control over the input-output vector of production. This control also im plies that the residual claimant has the managerial preroga- . tive. If the "firm" has the rights over any pure profits of 60 production, then it is something that has to be contracted for. Private property rights alone cannot guarantee any party being the firm. In an ex ante (pre-contractual) sense, there is no firm. Production can be viewed as the creation, transfer and destruction of private property. The property rights obtained by acquiring control over the productive activity will be non-zero even if profits are zero. This is illustrated easily. Consider a productive activity that uses H units of land and L units of labor, yielding Q units of output. The input-output vector is (Q,-L,-H). , The associated property vectors are (0,0,H) and (0,L,0) for the landowner and laborer, respectively. The market value of (Q,-L,-H) at prices (P,w,r) may be zero, but the production activity yields a property inequality, (0 ,0,H) ^ (0,0,H) + (Q,-L,-H). Thus, there is always some private gain, in a property-theoretic sense, that accrues to the firm. Extending the argument a bit further, it can be said that any future profits arising from a current production project does not have any future owner, if the current contract period is less than the life of the project. In this context it is clear why there is no indifference across agents as to who becomes the firm in the contract, but is determined by some characteristic as entrepreneurial 61 ability, or the ability to bear risk if the market value of the input-output vector is uncertain. Under sharecropping both parties jointly become the firm and it is precisely the residual claim that is shared. Suppose there is a sharecropping arrangement between a landlord and a tenant with some agreed upon share a*. It should be noted that a* is the rate at which the input-output vector is shared. Now think of a new technology that is available. This new technology can be used in a new production project that uses H units of land, L units of labor, yielding Q units of output annually for the next n years. The input-output vector is Y = (Q,-L,-H) and let the market price vector be P = (p,w,r), where w and r are the wage and rental rate, respectively. p is the output price. The profit is then given by: PY = pQ - wL - rH. Ignoring any maintenance costs, the present value of future land income, denoted by R, is: n rH 1 7 1 R = E — — + j=l(l+i)3 (1+i)n where i is the interest rate and T the terminal (or scrap) n 1 value of land. Denoting £ q- = A(n) , R can be j=l (1+i) written as: R = rHA(n) + T(l+i)_n 62 The discounted present value of the profit vector PY is: n V = E (PQ - wL - rH)/ (1+i)^ ° j=l = PYA(n) If the landlord hired labor and became the "firm", he would have obtained a discounted present value VH, given by: (3.3) VH = (pQ - wL)A(n) + T(l+i) n = (pQ - wL - rH)A(n) + rHA(n) + T(l+i)_n = PYA(n) + R V + R o If, on the other hand, the tenant leased the land and became the "firm", he would have obtained a discounted present value VT, given by: (3.4) VL = (pQ - rH)A(n) = (PY + wL)A(n) = PYA(n) + wIA(n) = V + W o 63 where W = wLA(n). Under sharecropping, both parties are residual claimants and a feasible sharing rate, 0 < a < 1, would have to share the present discounted value of the input-output vector V , such that = V^. If this could not be achieved then it is worthwhile for both parties to break away from the sharing contract and try to be the firm, individually. Consider an a which denotes the share of the tenant. Then VT = V„ implies: Li rl (3.5) (1-a)Y0 + R = aVQ + W or V - (W-R) o A number of possibilities arise. First, there is no guarantee that the a obtained from (3.5) is feasible, i.e. 0 < a < 1, Secondly, if some feasible a is obtained, it may not be at the ongoing rate a*. In either case, existing contracts will not suffice and they have to be renegotiated, leading to some new transaction costs. It illustrates the point made in the previous section, that new technology necessitates the reopening of contracts. Equation (3.5) is analyzed for the three cases that > may arise, namely W = R. 64 Case 1: W > R. a = 1 is impossible. a = 0 when Vq = W-R. 0 < a < 1 when VQ > (W-R). If Vq < (W-R), there is no feasible a. Case 2: W < R. a = 0 is impossible. a = 1 when Vq = R-W. 0 < a < 1 when V > (R-W). If V < (R-W), there is no feasible a. o o Case 3: W = R. a is always h. Thus, under the new technological possibilities, there exists a rationale for contracts to change. Any discussion of technological change must directly address the question of contract formation and the resultant transaction costs. Once the potential benefits of the new technology are weighed against the costs of trans actions, the issue of the adoption of new techniques and innovative behavior can be fruitfully discussed. 3.4 Concluding Remarks The framework of the previous section could be useful in empirical studies on the impact of new tech nology on contractual choice. Depending on whether W <<R, sharecroppers faced with a new technology may move over to new types of contracts . The value of W and R depends 65 on the time-horizon of planning, the interest-rate and the opportunity costs. It is useful to identify precisely the conditions under which sharecropping ceases to be feasible. It could provide a clue as to why in India, for example, sharecroppers in some regions adopted a new technology without changing contracts, while in some other regions new contracts were formed to adopt the same technology. This essay does not present a formal model, but suggests a way of analyzing technological change in the context of a private agent's cost-benefit computation. Though perfect competition was assumed to define a core allocation, it was not needed for the subsequent analysis. All that was required was any core allocation arrived at as the outcome of a principal-agent game. Bhaduri's monopolist landlord and tenant could be accommodated within the concept of a core allocation. Finally, the concept of transaction costs used in this chapter may be used to encompass a wide variety of non-economic factors. Cultural factors leading to differ ences in entrepreneurial ability, for example, would be reflected in differences in transaction costs. 66 CHAPTER 4 TRANSACTION COSTS, FOREIGN INVESTMENT AND ECONOMIC GROWTH 4.1 Introduction Development economists have traditionally pointed out that in order to accelerate economic growth, less developed countries can augment domestic savings by foreign aid and investment. The paucity of savings is explained by a time-preference pattern which favors consumption. It is as if a "big-push" is required to propel the economy out of a low-level trap to a path of self-sustaining growth. More skeptical writers have argued that aid and foreign investment may in fact, reduce domestic savings, and hence that the growth inducing effect of foreign savings has been exaggerated. Others have focused on the specific terms and conditions under which foreign savings become available, in particular the economic and political strings that are often attached. Marxists have argued that aid and foreign investment are instruments in the system of global imperialism by which developed countries dominate the less developed ones. A continua tion of aid and foreign investment perpetuates this 67 dependence, which in turn, they observe, acts as a fetter to indigenous change and development. Finally, looking around the world, there seems to be a growing hiatus between the rhetoric of self-reliance in economic development and the actual levels and trends of aid and foreign investment. In short, the experience, like the literature, seems extremely diverse. The issue is a politically sensitive one in which economic and political factors are closely intermingled, making it difficult to interpret and evaluate the evidence. One conclusion, however, clearly emerges, namely, that the effect of foreign investment on the overall growth trajectories of the economy does not operate merely through the savings constraint but also through changes in institutions, like the degree of rent-seeking behavior, the size and nature of the bureaucracy, and the innovative ingenuity and discipline of the work force. In other words, these institutions reflect costs that society incurs, arising from the complex configuration of contracts into which agents enter. The nature of these transaction costs are different across societies. Indeed, some economists like North (1981) have conjectured that transaction costs lie at the heart of institutional change and can be looked upon as the basis of a new economic history. A growing 68 body of literature has focused on the issue of transaction costs, but not much has been done in terms of incorporating it in the savings and investment framework. While some transaction costs arise within a firm in the production process, there are others that are external to it. Labor-shirking, for example, leads to transaction costs for a firm and is directly under its control. It can devise contracts or choose a technology that minimizes such costs. On the other hand, the same firm may have to pay a bribe to a dishonest bureaucrat to obtain a sanction to expand its capacity. In this case, the transaction cost incurred is beyond the control of the individual firm. The latter is an example of a transaction cost that society must incur (in this example, in terms of the future capital stock foregone) in maintaining the existing configuration of contracts. For the purposes of this paper, transaction costs are defined to be the additional costs incurred by investors, in terms of negotiating a contract with the state, to legitimize an expansion of productive capavity. The State's objective of maximizing rents for its agents may be in direct conflict with maximum economic growth. It is precisely this conflict that results in transaction costs. Using this notion of transaction costs, this paper analyzes the implications of foreign investment in a less- 69 developed country. Section 4.2 provides a rationale for the existence of foreign investment. Section 4.3 sets up a simple model of growth with foreign investment. Section 4.4 deals with policy implications. The final section presents some concluding remarks. 4.2 A Rationale for Foreign Investment Given the reservation with which some economists view foreign investment, and given the stated objective of self-reliance, it is puzzling to observe continued relaince on foreign investment in most less-developed countries. The apparent paradox can be resolved by considering transaction costs. Consider Figure 4.1, which depicts an economy with a given production technology and savings behavior. Implicit in the figure are the assumptions that workers do not save and that entrepreneurs save and invest a constant fraction of their profits. Let the savings of the entrepreneurs be srK, where s is the savings rate, r the rate of profit and K the stock of dK 1 dK capital. Then = srK, and g = — = sr, where g is the rate of growth of the capital stock. Thus, given s, there is a positive relationship between r and g. The factor- price frontier depicting the relation between r and the wage-rate w, is negatively sloped as drawn below. 70 A' w Figure 4.1 Transaction Costs and the Growth Rate The economy under consideration is a dual economy with a constant wage rate w*. Given w*, the rate of profit r = r* is obtained. Given r*, the rate of growth warranted by the production technology, the distribution of income and savings behavior, g* is obtained. However, in the presence of transaction costs associated with the conver sion of savings into investment, not all savings are 71 actually transformed into additions to the capital stock. Transaction costs shift the line OA to O'A'. The realized or actual rate of growth g° is lower than g*. The exis tence of transaction costs may explain why in many less developed countries growth performance has been poor, at least relative to expectations, inspite of fairly constant wages and generaly high rates of profit. The situation described in Figure 4.1 leads to a problem. Domestic investors realize that they are not being able to grow at their desired rates. Even if they recognize the presence and consequences of transaction costs, they are generally unable to do anything substantial toward reducing them. To resolve this problem, they would have to increase their profit rates. If wages cannot be lowered further (which would be true in the dual economy that was assumed), the only way out would be a search for a superior technology. The presence of transaction costs may inhibit any concerted effort to spend more on research and development programs. A faster and cheaper solution may be to seek foreign partners with whom they can enter into collaboration agreements so as to exploit the benefits of a superior technology. This is not to deny the fact that even without transaction costs domestic entrepreneurs would be interested in the latest in technology. Transaction costs merely reinforce this tendency. _______ 72] The government or the state may permit foreign investment because it can increase its rents (by taxes, licensing fees, etc.). Foreign investors may find it a suitable way to jump tariff barriers and trade restrictions, particularly in countries following an import-substitution strategy. Moreover, they stand to benefit from a cheap and assured labor supply. Private technical collaboration may thus emerge as a "marriage-of-convenience”. 4.3 Growth in the Presence of Foreign Investment Consider a one-good economy where part of the capital stock is controlled or owned by foreigners. The residue- is controlled by the domestic investors It is assumed that the foreigners possess a superior technology, and that the wage rate is given, so that producers have no control over it. The production function in each sector exhibits constant returns to scale, The production func tions can be written in terms of the average product of capital. Let the domestic and foreign sectors be denoted by the subscripts D and F, respectively. Y. So - = p = - = a - (£j) i = D,F / where Y is output, K the K . x x — x capital sotck and £ the labor intensity. It is assumed that an > a^, V 0 < £ < °°. F D Profit maximization implies: (4.1) a£, (£p) = w 73 (4.2) and (4.3) aF < V - £FaF < V = rF (4.4) w is the given wage rate and rp and the rates of profit. The four equations (4.1) to (4.4) determine the four unknowns and r^, i = F,D. It can also be shown that the foreign sector will choose a higher labor-intensity and earn a higher rate of profit. This is shown in the diagram below. It is assumed that the production functions satisfy the Inada conditions, namely: Both production functions start from the origin, implying that both factors of production are needed to produce any positive level of output. a± (0) = 0 a f (0) = 00 x ' a^ ( ' « > ) . = 0 . a . (00) =00 i = D,F x a!(£.) > 0 x x a '.'(£. ) < 0 x x 74 Figure 4.2 The Rate of Profit on Foreign and Domestic Capital The lines FF and DD have equal slopes, and the common slope is equal to the wage-rate. OF and OD are the rates of profit in the foreign and domestic sectors, respectively. Given the shapes of ap and aD, rp will always be larger than rD- Adjustments to equalize the rate of profit may be difficult to accomplish because if the foreign sector 75 is completely foreign owned, domestic capital cannot, by definition, become foreign capital. On the other hand, if the foreign sector is in collaboration with the domestic sector, there may be barriers to entry .into that sector, even though domestic capital would be likely to move into it. Another aspect of the solution to (4.1)... (4.4) that may seem surprising is that the foreign sector has a higher labor intensity. If the variable L is thought to be a composite dose of labor and intermediate inputs like raw materials, then one can think of the production function allowing substitution between capital and the composite inputs. In this situation, the ratio of non labor inputs to labor may be higher in the foreign sector. Let the total capital stock in the economy be K = Kd + Kp, where and Kp are the domestic and foreign capital stocks, respectively. It is assumed that the foreign and domestic sectors save a constant fraction of their profits, denoted by sp and sD, respectively, and that sF sD- Thus, the total savings of the private capitalists is given by: (4.5) SFrFKF + SDrD <K~ V There are transaction costs incurred in the conver sion of savings into investment. It is recalled that the essential nature of this cost is in terms of the capital 76 stock foregone. Though determined by the existing configuration of contracts and institutions, transaction costs are reflected as an additional cost on capital. The nature of the transaction costs is the leakage from savings accruing to government officials and bureaucrats. A part of this could find itself reconverted into savings. However, since the costs often arise from illegal transactions, it is highly plausible that most of it would be consumed. In the context of growth there is always a deadweight loss to the economy in terms of the leakage from savings. So, lacking a macroeconomic theory for the determination of transaction costs, it can be represented as a constant fraction of the capital stock. Transaction costs are then, (4.6) cK, 0 < c < 1 There is a government in the economy that taxes a constant fraction of output and spends it to facilitate private accumulation. The government does not own any capital stock. Government savings is then given by: (4.7) tY = t(apKF + aD (K-Kp)) where Y is total output and t the tax-rate. Obviously, 0 < t < 1. This specification abstracts from non-capital expenditures of the government. This is a simplifying 77 assumption. If non-capital expenditures are allowed, then t should be interpreted as the fraction of revenue that is spent on the building of overhead capital. The economy is engaged in international trade. Imports may be thought of as intermediate goods used in production. It is assumed that imports and exports are also constant fractions of output and that the exchange rate is given e = e. The trade deficit is then: (4.8) imY - enY = e(m-n)[apKp + aQ (K-Kp)] = b[apKp + aD (K-Kp)] where m and n are the propensities to import and export, respectively, and b = e(m-n). It is assumed 0 < (m,n) < 1. It is assumed that there are no taxes on trade. Alterna tively, it could also be assumed that the taxes on trade are equal to government consumption. Total savings in the economy is given by the sum of, (4.5) through (4.8). In equilibrium, investment is equal to savings. So, + b[apKp + aD (K-Kp)]. The rate of growth of capital stock 78 c + a_(t+b) + kF [sFrF - sDrD + t(ap - aD) + b(ap - aD)] or (4.9) U 1 dK , where g = — and a ^ K dt sDrD + aD (t+b)-c, the rate of growth achievable when the foreign share of capital kp = -g- == 0. It is noted that - s^r^ > 0 since r„ > r_, and F F D D F D sp sD> If b > 0, that is the economy has a trade deficit, then 3 > 0 unambiguously. 3 measures the effect the superior technology has on the growth rate of the total capital stock. (4.9) is an equilibrium condition and must hold at all moments of time. In (4.9), r. and a.(£.), 1 11 (i = D,F) are obtained from equations (4.1) to (4.4). g is highest when kp = 1. This reflects the positive impact the superior technology has on the rate of accumulation of capital. To obtain a steady-stage equilibrium value of foreign capital's share in the total capital stock, the growth rate of the labor force is assumed to be given exogenously: 8 = [sFrF - sDrD + t<aF 79 where g is the natural rate of growth. It is further assumed that the wage rate w is a market clearing wage rate. In other words, the labor market clears instantaneously to maintain full employment. Steady-state then implies that the economy-wide capital-labor ratio remains unchanged, or By definition, 0 < k„ < 1. In general, a and $ can — r — be of any sign. If domestic savings planned by private investors and the government is more than the transaction costs, and the economy has a trade-deficit, then a and 6 are both positive. Since this is fairly plausible in a developing economy, it is assumed that (a,3) > 0. Inspecting (4.11), it is noted that gn > a. A number of distinct possibilities arise. First, if gn £ a, then kF = 0. The economic meaning is clear. If the warranted rate of growth a, is at least as large as the natural rate of growth, then foreign savings are not required to augment growth because'in a steady-state the warranted rate of growth has to adjust to the natural rate. (4.10) g = a + Bkp = gn or (4.11) 80 The second possibility is that gn > a and gn - a >_ 6, g -a which implies —^— _> 1. Since kp is bounded by 1, the equilibrium value of kF is equal to 1. This means that if the growth-inducing effect of the superior technology is, at most, as large as the difference between the high natural rate of growth and the relatively low warranted rate of growth, then foreign savings has a strong growth- augmenting role and may completely dominate domestic capital. The only other possibility left is when g > a but gn~a <3. In this case 0 < kp < 1. If the growth- augmenting impact of foreign capital is larger than the difference between the warranted and natural rates of growth, then steady-state equilibrium may be reached before foreign capital completely dominates the economy. Here the role of foreign capital is essentially an auxiliary one, helping the economy reach a steady-state. However, as is clear from the second possibility, foreign capital can completely dominate a poor economy with a high rate of growth of population and low indigenous growth performance. In long-run equilibrium, the warranted rate of growth has adjusted to the natural rate and the labor market is in equilibrium with full-employment. However, from a developing economy's point of view, it is perhaps more pertinent to ask what would happen during disequili brium or temporary equilibrium in the intermediate run. 81 To throw some light on this issue, we slightly modify the above model. It is assumed that there is either unemployment or surplus labor in the economy, so that in the intermediate-run there is no upward pressure on wages. The economy can be thought to be described by the Lewis dual economy model. If wages remain constant, investors, by saving and investing more,can increasing the rate of accumulation and earn a higher rate of profit. In this sense, the purpose of stepping up the rate of accumulation is to earn a higher rate of profit. To proceed further requires an investment function derived from the profit- maximizing behavior of the capitalists. An investment function is derived for the foreign investors, Having an investment function for the domestic capitalists and letting the foreign investors be passive would not qualitatively change the results. However, the passivity of foreign investment would be hard to justify since it is at the request of domestic investors, foreign investment is present in the economy. Foreign capital is the dominant actor in the model, so the investment behavior of domestic capitalists is abstracted from even though how much they save is pertinent. The foreign investment function is derived by the method used by Lucas (1967) and incorporates the internal adjustment costs of the firm. Only here, these adjustment costs are 82 reinterpreted to incorporate the external transaction costs as well. Thinking of the foreign sector as a firm, its objective function may be written as: CO Max V(0) = I e“pt[F{L(t),K(t),I(t)} -wL(t) 0 - vl(t)]dt where p is the discount rate, w the wage rate, v the cost of investment (including the adjustment as well as trans action costs). F is the production function, homogeneous of degree one, satisfying FT >0, F^ > 0, FT < 0, F < 0, X j J \ X J - i J - i F < 0, F < 0, FTT = 0. F is concave. With some x \ J \ XX X i X restrictions on the derivatives of F, it can be shown that an optimal investment plan I(t) exists as long as the firm sets the marginal product of labor equal to the wage rate and the marginal product of capital equal to the marginal user cost. The investment function will be of the form: Ip(t) = Kp (t) G (w, v, p) , with g^j- < 0, g^- < 0 and < 0. Holding p constant, and w^.unchanging (by assumption) , investors will perceive that by investing more they can increase their rate of profit. Without loss-of generality, the investment function can be written as Ip (t) = Kp (t)G(r,v) with ^ > 0. 83 It may be recalled that r and g are positively related so that the foreign investment function can be written as: dKp 5C S t = H(9'v)' Hg = • F Thus, the higher is the rate of growth in the economy, the more foreign investors would like to grow. It is assumed that there is a minimum rate of growth, g . , that 3 ^ m m is necessary to induce a positive flow of foreign KF investment. From the definition of k„ = , It follows • « i1 AX k V F F K that rr~ - — - — , where the dot variable denotes a p K time-derivative. The transaction cost v reflects the adjustment costs foreign capital has to bear when it wants to increase its share arising from a higher visi bility and political vulnerability. This is written as: v = v (kp), v’ > 0. kp t— can now be written as: k (4.12) t— = H (g, v(k )) - g F * To analyze the investment behavior of foreign capital, the simplest explicit version of (4.12) is taken, as written below. 84 kF (4.13) , — = g - g . - vk„ k„ ^ ^mxn F F Equation (4.13) simply states that foreign investors choose to grow at a faster rate, the higher is the rate of growth in the economy, provided it is above some minimum rate, g . . The transaction costs incurred in ymin the process of investment increases with the share of foreign capital, making it more vulnerable to political threats like nationalization. For equilibrium, k^ = 0, or (4.14) kp (g - gmin - vkF) = 0. Going back t o ' ( , 4 .10), it has now to be interpreted as a contemporaneous relationship holding at every moment of time. (4.14) can now be written as: k„ = k„(a+3k_-g . - vk„) = 0 F F M F ymxn F' or (4.15) k = k (a - g • _ + (B-vJk-,) = 0 (4.15) is analyzed on the (k„,k_) phase space. Either r r • • k„ = k_ = 0, or k_, = a+3k„ - g . - vk_ = 0. Four cases F F F F ^mm F are possible: 85 (i) (ii) (iii) (iv) a > g • gain and V > 3 a > g • Jmin and V t 3 a < g • 3min and V > 3 a < g ■ and V < 3 Case (i) : k. F Figure 4.3 k_ has a positive intercept and a negative slope, F Two sub-cases arise. Either k^, = A and 0 < k^, < 1, or k„ = 1. Both solutions are stable. F Case (ii): k F kp = l Figure 4.4 k„ has a positive intercept and a positive slope. F Here k„ = 1 is a stable solution. Once again, complete F domination is possible. 87 Case (iii) k. F 0 Figure 4.5 F k„ has a negative intercept and a negative slope, F Here, the initial inducement is missing since a < and k„ falls to zero. k_ = 0 is stable, r r 88 Case (iv): k t F 0 Figure 4.6 kF has a positive slope and a negative intercept. As in Case (iii), the initial inducement is absent, and k_ = 0 is stable. k„ = A is unstable. An interesting F r possibility is that, if for some reason the initial kp is larger than A, then kp explodes to 1. This could be consistent with classical colonialism where a foreign government backs large doses of foreign investment leading to a complete domination of the economy, even though prior 89 to foreign penetration, the economy showed very little growth, i.e. a was very small. Since neither complete domination, nor a complete absence of foreign investment os observed, the case (i) solution, = A, is perhaps the most relevant one. This r solution to the intermediate-run equilibrium growth rate is a-g . , . i , m m (4.16) k_ = ---- 5— F v-3 4.4 Policy Implications For a discussion of policy options, a comparison between equations (4.11) and (4.16) is needed. In both cases, solutions which yield a share of foreign capital as a positive fraction are analyzed. From (4.11) gn'a F = gn " {sDrD + aD (t+b) In the long-run steady-state solution, when all adjust ments have been made in the economy, a reduction in transaction costs, c, and an increase in t or a devaluation (increase in e leading to an increase in b) would all lead to a fall in kp. This is a direct implication of the way transaction costs were modelled. There is no trade-off 90 between rapid growth and self-reliance. Given limits to an increase in t and an acceptable trade deficit, the burden of achieving self-reliance falls in the possibility of reducing c, through institutional change. In this context, the will of the state may diverge from the possibility of transformation. Looking at the temporary equilibrium value of kp in a Lewis-type of economy, where all adjustments have not been completed, policy implications seem just the opposite. From (4.16) : a-g . = mln F v-6 (5Drp + V t+*» - o - gmin_________ v {Sj,rp SDrD ^ ~ aj)) t t * (ap ~ a£)) ^ In this case, a reduction in c or an increase in t or e increases kp. There seems to be a trade-off between rapid growth and self-reliance, because any change that increases a, also raises k_,. However, it is obvious that an r increase in a is required for higher growth-rates. This dilemma is resolved when the additional transaction costs of the foreigners, v, is taken into account, A rapid increase in v with a simultaneous policy of reducing c and increasing t, can reduce kp and still permit a higher rate 91 of growth. During the transition phase of a dual economy, merely making it more difficult for foreign investors to operates does not guarantee rapid growth. A simultaneous policy of reducing domestic transaction costs is of crucial importance. There seems to be no trade-off between self-relaint growth and rapid capital accumulation. The crux of the proposition, however, depends critically on the ability of the state to reduce domestic transaction costs. Even if foreign investment persists, there is no substitute for the continued strive for efficiency by constant reductions of transaction costs. 4.5 Concluding Remarks This paper highlighted the transaction costs that arise in the conversion of savings to investment. Typically, they are high in less developed countries. It was used to provide an argument for the reliance on foreign investment. The objective of self-reliant growth can be achieved by a reduction of such costs. If the foreign sector is one of collaboration with the domestic sector, then the rapidity with which a superior technology can be absorbed is not only a function of the innovative ability of the domestic entrepreneurs, but also of the degree of monopoly foreign investment 92 enjoys over the technology. Typically less developed countries respect international patent rights, and foreign investment imposes restrictive clauses on technical collaboration agreements. Restrictive control over re search and development is extremely common and renders the absorption and adaptation of a new technology difficult. Transaction costs were treated as exogenous. In a wider context, transaction costs and technological change have to be tackled simultaneously. The government's role in changing transaction costs has to be analyzed. It is a difficult task, and the literature on this subject is in a somewhat nascent stage. Finally, the identifica tion and resolution of the transaction costs problem is a complex task. The state's role in this matter cannot simply be assumed. The issue can be summed up in the words of Douglass North (1981) : Relatively inefficient property rights threaten the survival of a state in the context of more efficient neighbors, and the ruler faces the choice of extinction or of modifying the fundamen tal ownership structure to enable the society to reduce transaction costs and raise the rate of growth. (p. 29) 93 CHAPTER 5 INTERNATIONAL TRADE AND GROWTH IN A DUAL ECONOMY 5.1 Introduction International trade theory, in both its classical and neoclassical versions, predicts that countries participating in trade stand to gain unambiguously, provided there are no barriers to trade. Each country exports the good in which it has a comparative advantage. Comparative advan tage is explained either by differences in technology or differences in factor endowments. It is in this context of the gains from trade, economists have observed that trade can be an useful "engine" of growth, during the development process. However, some writers have noted that the gains from trade in a static context have to be carefully examined, once economic growth is explicitly recognized in a dynamic context. The "immiserizing growth" literature concludes that rapid growth can reduce the price of developing countries' exports, leading to an adverse movement in the terms of trade, which in turn reduces the welfare of the exporting country. This will be highly probable if the foreign demand for the export good is inelastic. 94 Another branch of the literature that is critical of ~ trade as a vehicle of growth is the writings of the structuralists and neo-Marxists. Their argument runs as follows. The law of comparative advantage leads to a particular pattern of international specialization. The less developed countries are typically relatively well endowed with labor and hence export the more labor intensive commodities, namely the primary goods in the agricultural sector. As trade is opened up, resources get transferred to the more profitable primary goods sector. The non-export sectors shrink. Indeed, in the Ricardian model of trade, complete specialization in production is a highly likely outcome. Thus, for a developing economy, the realization of the gains from trade logically implies giving up all, or most of their non-export activities, in this case, manufacturing of indsutrial goods. Yet economic development and the structural change associated with it is supposed to increase the importance of the secondary or manufacturing sector. The presence of a small industrial sector makes the development process, from the developing economies' point of view, unequal. Moreover, these economies are heavily dependent on the developed coutries for capital goods and technology for the expansion of their productive capacities. 95 There is another aspect of this criticism that should be mentioned. The gains.from trade theorem is an aggregate one, and the issue of the incidence and internal (within the gaining country) distribution of the gains needs to be explicitly examined. If the gainers used their gains for current consumption, then trade would not add to the productive capacity in any way. Lewis (1977, 1980) has been aware of this problem implied by international specialization. He pointed out the asymmetry in the engines of growth in the developed and developing worlds. In the former, the drive for growth comes from the savings-investment mechanism and hence is internal. In contrast to this, in the developing economies the engine of growth is their exports of primary goods, and hence is external. To counter this dependecy, he argued for a build up of a domestic manufacturing sector that would maintain demand in the face of stagnant world trade in primary goods. Two recent papers, Krugman (1981) and Darity (1982) have tried to depict the unequal pattern of development in schematic form. A detailed criticism of Krugman and Darity is provided in the appendices to this essay. Krugman has shown, in a two-region world, at least one sector has to be completely specialized in either agricul ture or manufacturing. In his model the lack of an industrial sector is equated to uneven development. Yet ___________________ 96 the agricultural region produces its own capital goods, and can obtain manufactured goods through trade. The welfare implications of what he calls unequal development are not clear. It is more in keeping with the structural ist tradition to be able to show that trade may eventually lead to a situation where the productive structure of the less-developed region is in jeopardy. Darity, on the other hand, takes the Lewis argument to its long-run equilibrium and show that the Lewis pat tern of trade implies a persistence of the development gap. However, his result is dependent on the crucial assumption that the less developed region's own level of income is negatively related to its own rate of growth. This assumption is supported on the basis of an alleged upward pressure on wages and the need for rising "welfare measures" as growth proceeds. While there may some relevance of this argument, it implies that growth is, in a way, self-defeating. Growth in the developing countries, from the very beginning, is destined to taper-off. In contrast to this, in the developed regions the higher the level of income, the higher is the rate of growth. Thus, it is not so surprising that in the long-run the develop ment gap cannot be reduced. In this essay, a simple model is developed that highlights the issues discussed above, A general frame work is provided wherein the Krugman and Darity outcomes 97 constitute special cases. On the other hand, the standard predictions of neoclassical trade theory are also a possibility. The long-run fate of the developing countries, in terms of which result is actually arrived at, depends crucially on the behavior and preferences of the elites of these countries. It is shown that the structuralist implication of trade is not inevitable. Much depends on what the participants make of the opportunities that trade affords. This essay is organized into four parts. Section 5.2 describes the basic model. Section 5.3 develops the dynamics. Section 5.4 contains some concluding remarks. 5.2 The Model Consider an economy with two sectors — agriculture and industry. The agricultural sector produces a non food raw material as well as food. It is assumed that the economy has a comparative advantage in the raw material which it exports. It is also assumed that this economy is relatively well endowed with labor. This, and the law of comparative advantage, imply that the primary export good is the most labor intensive good. The industrial sector produces a capital good which is used as an input in the production of both agricultural goods. It is assumed that the real wage rate is fixed in terms of food. It is also assumed that the industrial 98 sector output is demand-determined. The price of food is taken as the numeraire. The above implies that for this model the terms of trade between food and the industrial good is constant. This is a simplifying assumption. As will be shown later on, making the price of the industrial good endogenous does not alter any qualitative results. Moreover, keeping the domestic terms of trade constant helps highlight the impact of international trade. Within the agricultural sector, the raw material export and food are substitutes. Any increase in the output of one reduces the output of the other. Before describing the model, a list of notations used is provided below. X output of the raw material (export) F output of the food sector the amount of the industrial good K used in the production of X K F the amount of the industrial good K used in the production of F N X employment in the X sector N ,F employment in the F sector M amount of imports P X price of X in terms of food P. M price of M in terms of food P K price of K in terms of food 99 W : the wage-rate in terms of food s : the tor surplus in the agricultural sec- after the wage bill is paid K : total production of the industrial good The model can be described by the following equations: (5.1) X = X(W, P X ' PK} (5.2) F = F (W, P X' PK) (5.3) NX = NX (W, px' V (5.4) nf = nf (W, PX' V (5.5) KX = KX (W, PX' PK) (5.6) F K = kf (W, PX' V (5.7) W = W (5.8) PK = P K Equations (5.1) and (5.2) are the supply functions of X and F. X responds positively to an increase in P^, while F declines when (5.3) and (5.4) are 3NX _ „ 3N , „ 3PX > 3PX " °- the industrial good 9 X 9 F rises. This implies - r — - > 0 and ^ < 0. dPX X the labor demand.functions, with (5.5) and (5.6) denote the demand for V F 3F 3 k in X and F, with > 0 and ^ <0. X X 100 (5.7) and (5.8) determine the wage rate and the price of the capital good. The surplus in the agricultural sector can be written as: (5.9) S = PXX + F - W(NX + NF) S accrues as income to the agricultural landlords who have to spend a part of it in purchasing the industrial good and importing goods from abroad. Setting their income equal to expenditure, the following is obtained: (5.10) S = PmM + Prk where (5.11) K = KX + KF To abstract from the effects of a trade deficit, it is assumed that trade is balanced. So, (5.12) PXX = P ^ Finally, to close the model, it is assumed that the economy is small. So, px (5.13) = a M Equations (5.1) to (5.13) describe a model in 13 unknowns. Assuming an equilibrium to exist, the implica- tions of the model can be studied. 101 The gains from trade imply that an improvement in the terms of trade should increase the surplus in the agricul tural sector. This is assured if the price elasticity of the supply of X is greater than the cross-price elasti city of F. This assumption is made, noting that the elasticities of supply of food are usually low in developing countries. An increase in a will lead to an increase in Pv or a fall in PM . A rise in P^ will stimulate exports and consequently X will rise. If PM falls, imports will rise and the expenditure on imports will rise if the elasticity of demand for imports is greater than one, which is plausible in less developed countries. In both cases, X will rise. Thus, the value of exports P X goes up. A Output of food falls. The wage bill rises from the assumption that the export good is more labor intensive. If the increase in the value of exports more than offsets the fall in F and the increase in the wage bill, then S will increase. Note that here the "gains" from trade benefit workers in agriculture as well as the producers. dS It is assumed that this holds, i.e. 3— > 0. da Equations (5.9), (5.10) and (5.12) together imply: (5.14) F - W(NX + NF) = PrK If the wage-bill rises and F falls, the demand for fertilizer falls. This is depicted in (5.14), which shows 102 the food surplus on the left hand side that is used to demand the industrial good. So output of the industrial good unambiguously falls. This emanates from an income effect on demand. If there was an independent supply curve to endogenously determine Pv, a fall in demand (from J \ a fall in income) would lead to a fall in PR and the equilibrium output of K. The shape of the supply curve would determine the magnitude of the reduction Faced with a falling demand, the industrial sector shrinks. So far, nothing has been said about the nature of the import good M. If M is a consumption good, then an improvement in the terms of trade will ultimately affect the production of both X and F via the reduction in the size of the industrial sector. If, on the other hand, M is a capital good, then too, the problem arises of finding food from some source. The composition of M seems to be crucial in the long-run. It requires some conscious policy to sustain growth. This is exactly the point made by Lewis. The following section develops the dynamic implications of the model. 5.3 The Dynamics To explore the long-run equilibrium implied by the model, the economy described is taken to represent a less developed region called the South. The South trades with 103 ------ — ------------------- . — .... — .— — ___________________i a developed region called the North whose engine of growth is its own savings-investment mechanism.. A higher income leads to higher savings and investment, leading to a higher rate of growth. Its growth is inde pendent of the South. So, (5.15) gN = gN (KN), gj > ° where gN is the rate of growth in the North and its capital stock. In the South, higher income also leads to higher rates of growth. The higher is the capital stock, the higher is its income. So far, the logic of its growth is similar to that of the North. However, trade also affects its growth. If higher income in the North improves the terms of trade faced by the South, then as the model suggests, the South's industrial structure and capital stock gets reduced. This is the dependency effect. The North can be incorporated into the model by rewriting (5.13) as: px (5.16) ^ = f(Kn) , f* > 0 M Let gc denote the rate of growth in the South. Then the above arguments can be summarized as: (5.17) gs = gs (KN, Kg), 0, g ^ > 0 104 where Kg is the capital stock in the South, gg and g( 3g "9g< 3K and N 3K, “iVS The long-run equilibrium of the system (5.15) and (5.17) can be depicted on the (KN, Kg) phase space. Several cases are possible. Case (i): If g„ < 0 dominates, i.e., the interna- 1 tional dependency effect is very strong, the gg = 0 locus has a negative slope. ’N 0 is a vertical line, since does not depend on Kg. This is shown in Figure 5.1 (a). KS A 2 I I o / t ? 9s = / / _________ K N Figure 5.1'(a) Strong Dependency Effect 105 E is a stable interior solution. If E lies to the right of the 45° line, then the North ends up with a larger capital stock and hence a bigger industrial sector. This is predicted by-comparative advantage. It is also analogous to the Darity result where the "gap" is not eliminated. Darity's result, however, is perfectly consistent with the law of comparative advantage, since the gains from trade say nothing about levels of income or the size of the industrial sector being equalized. As can be noticed by now, much depends on the position of the gg = 0 locus. If trade can be utilized to augment, through imports, the capital stock of the country, then gg = 0 would move upwards. In other words, for a given K^, Kg would be higher, the higher is capital imports. Indeed, the logical possibility of an intersec tion (stable equilibrium) on the left of the 45° is also a possibility, where the long-run comparative advantage has been reversed, the South ending up with a higher capital stock. If, instead of importing capital goods, all or most of the imports are consumption goods, then the gg = 0 locus will move downwards. This may be ruinous for the economy because without the industrial capital good, agricultural output itself cannot be sustained. This is the case that the radical authors argued and Krugman tried 106 to formalize. This case is depicted in Figure 5.1(b). K S kN 0 Figure 5.1 (b) Strong Dependency Effect - The Krugman Case Case(ii): If gr > 0 dominates, i.e., the internal 2 engine of growth is stronger in the South and trade is relatively less important, a stable solution exists. As before, it can lie on either side of the 45° line. This is shown in Figure 5.2. 107 Figure 5.2 Dominant Internal Effect Case (iii): Instead of assuming that one of the two opposing forces dominates gg, it can be assumed that ultimately one of the two forces overpowers the other. Two situations are possible. First, as depicted in Figure 5.3(a), the dependence effect initially dominates, but ultimately, the internal engine overcomes the depen dence effect. It may be noted that some of the newly industrialized countries were ex-colonies, where the dependence effect would be extremely strong. In Figure 108 5.3(a), since both solutions are locally stable, a movement from E2 to would require some planning and something in the nature of a "big-push". Figure 5.3 (a) The Ultimate Domination of the Internal Effect The final possibility is where the dependence effect ultimately dominates, as shown in Figure 5.3(b). Once again, both solutions are stable. 109 A *1 g N ° t-> % <-* g s = o ^ Figure 5.3(b) The Ultimate Domination of the Dependency Effect It may be concluded that unequal development, though a distinct possibility is not inevitable. 5.4 Concluding Remarks The results of the previous sections critically depend on what is done with the gains from trade. The need for some planning of the import composition was also noted. The problem of unequal development and dependence must be analyzed in terms of the influential elites in the 110 underdeveloped countries, their behavior, their preferences and the institutional framework that makes them what they are. The essence of the structuralist proposition implies that there must be a critical size of the industrial sector before growth can be self-sustained. Indeed, this is also what authors like Lewis or even Rostow (1961) tried to say. Viewed in this light, the absence or near absence of an industrial sector makes an economy dependent on some external source for growth, beyond the realm of its own control. A caveat should also be added to the possibility of export-led growth. Imports of new industrial capital goods usually embody a new technology. Its efficient use, absorption and adaptation is a tricky issue and deserves more attention on its own merits. It is well-known, for example / that the importing of a tractor is not a sufficient condition for agricultural productivities to rise. New technology may require new types of transactions leading to someone having to bear the transaction costs. This must be explicitly recognized in any meaningful strategy of rapid development. Finally, the model described for the South should be interpreted as a model of the least developed region in the world economy. This is because the external demand 111 for the South's industrial good was not considered. A consideration of external demand could imply that, as the industrial sector faced a falling domestic demand, it could turn to an external market. The assumption of a constant PR would then have to be modified. If the initial was higher than the world price, the industrial sector would have to increase supplies so as to reduce the price PK, and try to sell the output abroad. If this possibility arose, leading to a fall in P„ through supply effects, the IV domestic quantity demanded would also tend to increase. The possibility of the dependency effect dominating the economy would substantially diminish. In this context, not only would the composition of imports and the behavior of the landlords be important, much would depend on the ability of industrialists to constantly reduce costs and search for new markets. 112 APPENDIX 5.1 Krugman (1981) presents a formalization of uneven development. He analyzes two symmetric regions that can produce both a manufactured and an agricultural good. An important specification is that the agricultural sector produces the capital goods in the economy. There are economies of scale in industry. As growth proceeds, the world output of manufactured goods increase, leading to a fall in its price. If world output increases via an increase in accumulation in the North, the price of manufactures would fall leading to a fall in the rate of profit and hence also the rate of investment, since all profits are reinvested in the South. One region's growth always hurts the other region. Growth in each region produces two opposing effects. The industrial price drops, profits fall and investment declines. However, capital accumulation, thanks to external economies, reduces costs and increases the rate of investment by stepping up the rate of profit. Krugman assumes that the former is more important. In this world, any interior solution to the long-run equilibrium in (KN, Kg) space is unstable. (K^ and Kg are the industrial capital stocks in the North and South, respectively). The effect of this instability is bounded _____________: ___ 113 by the maximum capital stock that can be attained, given the labor force, in each region. The implication of this type of solution is that both regions cannot be unspecialized simultaneously. Where the global economy comes to rest is determined by the initial conditions (primitive accumulation is interpreted by Krugman). The region that ultimately becomes underdeveloped is either completely specialized in agriculture, or has a small manufacturing sector. In either case, the region produces its own capital goods (in the agricultural sector) and hence is not constrained by trade to attempts to increase its productive capacity. Even when manufactured goods are not produced at all in this region, the workers' consumption (the workers are the only consumers) of the manufactured good is not affected, since by assumption their demand is price inelastic. In the absence of an industrial structure of its own, the Southern workers' demand is satisfied from imports. The only adverse effect is that of a declining terms of trade. The capitalists in the underdeveloped region live with falling profits and hence decelerating investment in the manufactured good. Krugman's results imply two unusual features. For him, free trade always hurts both regions through declin ing industrial prices, as growth proceeds. There are 114 no gains from trade. Price intervention (to prevent prices from falling) would benefit all capitalists everywhere. The workers, anyway,' do not care about prices (their demand being perfectly inelastic). This possibility is not addressed by Krugman. For Krugman, underdevelopment is the absence of a large industrial sector. Even when this economy can consume all the industrial goods it wants, it is still underdeveloped. This implication goes against the radical critique. Trade produces a one-way dependency, adverse for the less developed region. The nature of this dependency, if allowed to continue into the long-run, would ruin all productive activities in the underdeveloped world. Krugman fails to incorporate this very important aspect of the radical critique, even though his model is illuminating in some other aspects. The above essay tries to correct Krugman's short coming. 115 ... t Appendix 5.2 In Darity's (1982) model the North (developed region) has its own internal engine of growth. The growth rate of income per capita, is positively related to the investment per head. Investment is positively related to per capita income in the usual Keynesian fashion. However, investment is also determined by the wage-rate which is negatively related to it. The wage-rate, in turn, is positively related to income. Darity assumes that ultimately the retarding effect of rising labor cost reduces the rate of growth as income rises. However, the direct influence of income on the growth rate is always positive. For the South (the less developed region) there are two opposing influences on growth. First, Northern growth is positively related to Southerh growth. As the South's own income increases, it always reduces growth in that region. The South's growth rate function is: gS = m(gn , ys), m^ > 0 , m2 < 0 . n s where g , g are the per capita rates of growth of income g in the North and South. y is the Southern income per capita. 116 Two objections can be raised against this formulation. First, the dependency effect to be positive implies that Northern growth is beneficial to the South. This, one can argue, goes against the spirit of the structuralist and Marxist critiques. Northern growth is somehow seen to be adverse for Southern development. It is in that sense development of the North and underdevelopment of the South is seen to be part of the same global growth process. Secondly, the South's own income adversely affects its own rate of growth, i.e. m2 < 0. Yet this is the South's own internal engine of growth. In Darity's formulation, the South, if it wants to be independent, must reduce dependency. It immediately follows then, that the South is doomed, with m2 < 0 dominating. Given his formulation, dependency for the South is its best solution. This is in direct contradiction to the connotation of "dependency" in the structuralist critique. Moreover, the two internal engines of growth are asymmetric. For the North, a rise in its own income positively feeds back on the rate of growth. For the South, it is exactly the opposite. Darity provides no explanation for this. Lewis, and the proponents of the structuralist critique, had perhaps the opposite sign-configuration of m^ and m2 in mind. It is trade with the North that may, 117j through the terms of trade and a shrinking industrial sector, adversely affect the South. To counter this, a conscious development of industry (to sustain its own growth) must be attempted by the South. The above essay attempts to formalize this interpre tation of Lewis. 118 CHAPTER 6 CONCLUSION As mentioned at the outset, the task of these essays is to reconcile the Lewis framework to some of the major criticisms made against it. Since the criticisms could be broadly categorized into those emanating from the neo classical school of thought, and those arising from a structuralist or neo-marxist perspective; a successful accomplishment of this task would also serve, at least to some extent, to help bridge the gap between these two schools. In this context, a clarification is in order. In trying to reconcile the two schools of thought, the purpose has not been to prove that anything might happen, but only to question the inevitability of some of the results that have been claimed by the various critics. The task is to investigate when, and under what specific conditions, the main propositions of the alternative schools of thought would hold. This was done for a labor- surplus, dual, Lewisian economy. Even though the investigations were not carried out in terms of a single analytical model, the main characteristics of the Lewisian economy were retained throughout. Four vitally important and logically connected issues were taken up, namely, the 119 internal terms of trade, agricultural technology, foreign investment and international trade. A brief summing up of the results is presented below. A crucial argument of Lewis was that during the "surplus" phase of a dual economy prices and wages would remain constant. Under this circumstance rapid capital accumulation could proceed before the phase of "shortage" arrived, and the market economy takes over the function of efficiently allocating scarce resources. The neoclassical response was that as development took place, the rising demand for food would drive up the relative price of food and wages. The existence of the "surplus" phase was questioned. The implication was that from the beginning the laws of the market economy held sway and hence, analysis needed to be carried out in terms of allocative efficiency. On the other hand, the neo-Marxist critique had overemphasized the supply rigidities in food produc tion. The price of food was bound to rise, thwarting the Lewisian possibility of rapid growth. The upward pressure on food prices would be inevitable, stemming from the institutional framework of labor-surplus agricultural sectors. The monopolistic or semi-feudal power of the landlords could be removed only by a social revolution. Otherwise, the Lewisian economy would bound to stagnate. In both cases, although for quite different reasons, the 120 terms of trade would invariably move against the industrial sector. In Chapter 2, both supply rigidities and demand constraints were incorporated. The temporary equilibrium in the economy could be one that is characterized by either excess supply or excess demand. Full employment is also a possibility. Comparative statics exercises reveal that when wages increase or the agricultural sector changes, the terms of trade would not necessarily move against industry. In general, the comparative statics results depend on the size of the marketed surplus and the values of the marginal propensities ,to consume by the different economic groups. In the agricultural sector the availability of a new technology was not sufficient to ensure that it would be adopted. This was demonstrated in Chapter 3. Whether a new technology would be adopted, given the existing institutional structure, or whether new institutions and new contracts had to be devised, depended on the trans action costs involved and the perceived opportunity costs. Neither is the mere availability of new technology sufficient to guarantee its adoption, nor is institutional reform always necessary. The conditions under which each case would be true are derived. 121 The issue of raising productivities and the rate of accumulation was noted to be crucial for the industrial sector. Neoclassical economists have maintained that foreign investment can be a powerful vehicle to this effect. The Lewisian economy with foreign investment would be unambiguously better-off than the same economy without foreign investment. Neo-Marxists, on the other hand, have persistently argued that foreign investment is a tool of domination of the developed countries that is used to gain complete control over the modern sectors of the less developed countries. The Lewisian economy, being open to foreign investment, would be inevitably reduced to a subservient status. In Chapter 4 a rationale was provided, in terms of the presence of transaction costs, for the existence of foreign investment in the developing countries. It was shown to be possible that, as accumulation takes place, the share of foreign capital could approach unity. This was the thrust of the propositions contained in the modern theories of imperialism. However, it was also noted that, whatever be the share of foreign capital in the total capital stock, its contribution to economic growth was unambiguously positive. If the objective of the economy were to maximize the rate of growth, foreign investment would be an extremely useful tool. Instead, 122 if the objective were to achieve self-reliant growth, then ways and means would have to be devised to reduce trans action costs. Neoclassical economists have also maintained that the Lewisian economy, by participating in international trade can unambiguously benefit from the gains from trade and international specialization. Neo-Marxists have sub scribed to an opposite view. Trading with the rest of the world, particularly the developed capitalist countries, would reduce the Lewisian economy to a subservient status and only help to perpetuate stagnation. Trade has not been seen as an engine of growth for the developing countries. Instead, it was seen as an unequal exchange mechanism favoring the already developed countries. Chapter 5 attempted to reconcile these opposing views. The presence of gains from trade was found to be insuf ficient for growth. The crucial issue was, and is, what is done with the gains from trade. If the gains are frittered away on the consumption of luxuries, then it could result in the type of dependency the neo-Marxists have mentioned. However, there was nothing inevitable about economic stagnation resulting from international trade. The search for a paradigm of economic development, a truly general theory, is far from complete. The origin 123 for this search is by no means confined to the Lewis model. However, the richness of Lewis' writings offers an attractive starting point. A very small beginning was attempted. If, as a result, the widening gap between the two schools of thought were reduced even marginally, that would be considered a step forward in the translation of economic theories into basic bread and butter propositions for the world's masses of extremely poor people. 124 BIBLIOGRAPHY Amano, M. (1980). "A Neoclassical Model of the Dual Economy with Capital Accumulation in Agriculture", Review of Economic Studies, Vol. XLVII(5), No. 150, pp. 933-944. Amin, S. (1979). Unequal Development, Delhi, Indian Edition, Oxford University Press. Arrighi, G. (1970). "Labor Supplies in Historical Perspective. A Study of the Proletarization of the African Peasantry in Rhodesia", Journal of Development Studies, Vol. 6 , No. 3, pp. 197-234. Bagchi, A. (1982). The Political Economy of Underdevelop ment, Cambridge, Cambridge University Press. Baran, P. (1962). The Political Economy of Growth, New Delhi, People's Publishing House, Indian Edition. Bauer, P. (1956). "Lewis' Theory of Economic Growth", American Economic Review, Vol. XLVI, No. 4, pp. 63 2- 641. Bell, C. and Zusman, P. (1976). "A Bargaining Theoretic Approach to Cropsharing Contracts", American Economic Review, Vol. 66, No. 4, pp. 578-588. Bhaduri, A. (1973). "Agricultural Backwardness under Semi feudalism", Economic Journal, Vol. 183, No. 329, pp. 120-137. Bhardwaj, K. (1979). "Towards a Macroeconomic Framework for a Developing Economy: The Indian Case", Manchester School Of Economic and Social Studies, No. 3, pp. 270-302. Bose, S. (1968). "Optimal Growth and Investment Alloca tion", Review of Economic Studies, .Vol. XXXV(4), No. 104, pp. 465-480. Brewer, A. (1980). Marxist Theories of Imperialism: ,A Critical Survey, London, Routledge and Kegan Paul. 125 Chackravarty, S. (1977). "Reflections on the Growth Process in the Indian Economy", in Wadhva, C. (Ed.) Some Problems of India's Economic Policy, Second Edition, New Delhi, Tata-McGraw Hill, pp. 113-136. Chichilinsky, G. (1981). "Terms of Trade and Domestic Distribution, Export-led Growth with Abundant Labor", Journal of Development Economics, Vol. 8 , No. 2, pp. 163-192. Clower, R. (1969). "The Keynesian Counter-Revolution: A Theoretical Appraisal", in Clower, R. (Ed.) Monetary Theory, Penguin Modern Economics Readings, Harmonds™ worth, Penguin Books, pp. 27 0-297. Darity, W. (1981). "The Simple Analytics of Neo-Ricardian Growth and Distribution", American Economic Review, Vol. 71, No. 5, pp. 978-993. _________________ (1982). "On the Long-run Outcome of the Lewis-Nurkse International Growth Process", Journal of Development Economics, Vol. 10, No. 3, pp. 271-278. David, P. (1975). Technical Change, Innovation and Economic Growth, Cambridge, Cambridge University Press. Dixit, A. (1968). "Optimal Development in the Labor Surplus Economy", Review of Economic Studies, Vol. XXXV(l), No. 101, pp. 23-34. _________________ (1969). "Marketable Surplus and Dual Development", Journal of Economic Theory, Vol. 1, No. 2, pp. 203-219. _________________r (197 0) . "Growth Patterns in a Dual Econo-? my", Oxford Economic Papers, Vol. 22, No. 2, pp. 229- 234 . _________________ (1973). "Models of Dual Economies", in Mirlees, J. and Stern, N. (Eds.) Models of Economic Growth, New Delhi, Macmillan, pp. 325-352. Emmanuel, A. (1972). Unequal Exchange, A Study of the Imperialism of Trade, New York, Monthly Review Press. Fei, J« and Ranis, G. (1964), Development of the Labor Surplus Economy, Illinois, Irwin. 126 Fei, J. and Ranis, G. (1966). "Agrarianism, Dualism and Economic Development", in Adelman, I. and Thorbecke, E. (Eds.) Theory and Design of Economic Development, Baltimore, Johns Hopkins Press, pp. 3-44. Findlay, R. (1978). "Relative Backwardness, Direct Foreign Invesmtnet, and the Transfer of Technology: A Simple Dynamic Model", Quarterly Journal of Economics, Vol. XCII, No. 1, pp. 1-16. Frank, A. (1978). Dependent Accumulation and Underdevelop ment , London, Macmillan. Gersovitz, M. (1974). "Aggregate Demand, The Wage Gap and Unemployment in LDCs", Journal of Development Econom ics, Vol. 1, No. 3, pp. 235-246. Griffin, K. (1969). Underdevelopment in Spanish America, London, Allen and Unwin. Guha, A. (1969). "Accumulation, Innovation and Growth under Conditions of Disguised Unemployment", Oxford Economic Papers, Vol. 21, No. 3, pp. 360-372. Harris, D. (1978). Capital Accumulation and Income Distribution, Stanford, Stanford University Press. Hirschman, A. (1967). Development Projects Observed, Washington, D.C., The Brookings Institution. Hornby, J. (1968). "Investment and Trade Policy in the Dual Economy", Economic Journal, Vol. LXXVIII, No, 309, pp. 96-197. Jorgenson, D. (1961). "The Development of a Dual Economy", Economic Journal, Vol. LXXI, No. 282, pp. 309-334. : ______________ (1967). "Surplus Agricultural Labor and the Development of a Dual Economy", Oxford Economic Papers, Vol. 19, No, 3, pp. 2 81^-306. Kalecki, M. (1976). Essays on Developing Economies, New Jersey, Humanities Press. Kotwal, A. (1981). "Choice of Contracts and Technological Change in Indian Agriculture", University of British Columbia, Disccusion Paper 81-32. 127 Krugman, P. (1981). "Trade, Accumulation and Uneven Development", Journal of Development Economics, Vol. 8, No. 2, pp. 149-162. Leeson, P. (1979). "The Lewis Model and Development Theory", Manchester School of Economic and Social Studies, No. 3, pp. 196-210. Lewis, W. (1954). "Economic Development with Unlimited Supplies of Labor", Manchester School of Economic and Social Studies, Vol. XXII, No. 2, pp. 131-191. _________________ (1958). "Unlimited Labor: Further Notes", Manchester School of Economic and Social Studies, Vol. XXVI, No. 1, pp. 1-32. _________________ (1972). "Reflections on Unlimited Labor", in Luis Engenio Di Marco (Ed.) International Economics and Development Essays in Honor of Raul Prebisch, New York, Academic Press, pp. 7 5-96. _______ . ______ (197 8). The Evolution of the International Economic Order, Princeton, Princeton University Press. _________________ (1979). "The Dual Economy Revisited", Manchester School of Economic and Social Studies, No. 3, pp. 211-229. « _________________ (198 0). "The Slowing Down of the Engine of Growth", American Economic Review, Vol. 70, No, 4, pp. 555-564. Lucas, R. (1967). "Adjustment Costs and the Theory of Supply", Journal of Political Economy, Vol. 75, No. 4, pp.321-334. McIntosh, J. (1975). "Growth and Dualism in Less Developed Countries", Review of Economic Studies, Vol. XLII(3), No. 131, pp. 421-434. . Mitra, A. (1977). Terms of Trade and Class Relations, London, Frank Cass. Newbery, D. (1975). "Tenurial Obstacles to Innovation", Journal of Development Studies, Vol. 11, No. 4, pp. 263-277. Niho, Y. (1976). "The Role of Capital Accumulation in the Industrialization of a Labor Surplus Economy", Journal of Development Economics, Vol. 3, No. 2, pp. 161-170. 128 North, D. (1981). Structure and Change in Economic History, W.W. Norton and Company. Owen, R. and Sutcliffe, B. (Eds.) (1972). Studies in the Theory of Imperialism, London, Longman. Palma, G. (178). "Dependency: A Formal Theory of Under development or a Methodology for the Analysis of Concrete Situations of Underdevelopment", World Development, Vol. 6, No. 7/8, pp. 881-924. Rakshit, M. (1982). The Labor Surplus Economy: A Neo- Keynesian Approach, Delhi, Macmillan. Rostow, W. (1961). Stages of Economic Growth: A Non- Communist Manifesto, Cambridge, Cambridge University Press. Schultz, T. (1964). Transforming Traditional Agriculture New Haven, Yale University Press. Taylor, L. (1979). Macro Models for Developing Countries, New York, McGraw Hill. Taylor, L. and Bacha, E. (1976). "The Unequalizing Spiral: A First Growth Model for Belindia", Quarterly Journal of Economics, Vol. XC, No. 2, pp. 197-218. Tendler, J. (1975). Inside Foreign Aid, Baltimore, Johns Hopkins University Press. Wallerstein, I. (1979). The Capitalist World Economy, Cambridge, Cambridge University Press. Wellisz, S. (1968). "Dual Economies, Disguised Unemploy ment and the Unlimited Supply of Labor", Economica, Vol. XXXV, No. 137, pp. 22-51. Wilber, C. (Ed.) (1973). The Political Economy of Development and Underdevelopment, New York, Random House. Zarembka, P. (1970). "Marketable Surplus and Growth in the Dual Economy", Journal of Economic Theory, Vol. 2, No. 2, pp. 107-121. 129

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Money, credit and economic growth in fragmented economies

PDF

Irrigation institutions and technical change: A case study from South India

PDF

Countertrade: A theoretical and empirical investigation

PDF

Patronage, dependence and debt in peasant agriculture

PDF

The price of oil and OPEC behavior: A utility maximization model

PDF

The economics of dowry

PDF

Three essays on the determinants of firm structure and contractual choice

PDF

Life-cycle demographic decisions and the economic well-being of single parents in the United States, 1964-1989

PDF

An empirical investigation of industry conduct under information sharing: The case of the gypsum industry

PDF

Immiserizing effects of aid-induced price distortions

PDF

The influence of industrial pressure groups within the People's Republic of China

PDF

Exchange rate volatility and regular dynamics of the exchange rate in general equilibrium models of exchange rate determination

PDF

Export development of Iran, 1900-1982

PDF

Diversified expectations in the foreign exchange market

PDF

Two essays in political economy of education and economics of immigration

PDF

An economic study of the social costs and benefits of a no-fault automobile accident victim indemnification system

PDF

Exchange policy: A research on Peru's rate of exchange and external balance

PDF

An economic model of the security motive in the demand for children: An application to Malaysia

PDF

Voluntary overcompliance of environmental standards: Theory and empirical evidence

PDF

Complex economic growth

Asset Metadata

Creator Sinha, Anup Kumar (author)

Core Title Essays on some Lewisian themes of economic development

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Economics

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Economics, General,OAI-PMH Harvest

Language English

Advisor Nugent, Jeffrey B. (committee chair), Cooper, Eugene (committee member), Elliott, John E. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c20-267103

Unique identifier UC11260097

Identifier DP23325.pdf (filename),usctheses-c20-267103 (legacy record id)

Legacy Identifier DP23325.pdf

Dmrecord 267103

Document Type Dissertation

Rights Sinha, Anup Kumar

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA