TWENTy-FIVE YEARS OF SWEDISH ECONOMIC RESEARCH ON LATIN AMERICA : 1969-1993

Over the course of the past twenty-five years, Swedish social scientists have cultivated a keen interest in Latin America. This is particularly true of Swedish economists. The generation born in the 1940s has devoted considerable attention to Latin American issues since the early and mid-1970s. So have a number of Latin American scholars living and working in Sweden since the establishment of military rule in Chile,Argentina and Uruguay around that same time. Since then a steady flow of works has seen the light of day. However, no attempt has ever been made to survey this particular body of literature. Hopefully, the present survey will serve to bridge that gap. I have chosen to employ a 'broad' definition of 'economic' research. This is only natural in the present context, in view of the fact that LatinAmerica is a continent of developing nations, and in development economics in general the dividing lines between' economics', 'economic history' and' economic geography' are rather blurred. Interdisciplinary approaches are much more common here than in other fields of economic science. This implies no particular value judgment on my part. Whatever one may think, it is a reality that should be reflected in the survey. Business administration studies, on the other hand, have been interpreted as falling outside the scope of the present survey,l and so have, with a single exception, studies carried out within the disciplines of political science, social anthropology and sociology.2 I have also attempted to exclude works which obviously overlap as well as mere translations of existing pUblications. The works surveyed herein have been grouped into the following, somewhat arbitrary, categories, which, with one or two exceptions, correspond more or less to the headings one normally finds in textbooks on development economics:


TWENTy-FIVE YEARS OF SWEDISH ECONOMIC RESEARCH ON
LATIN AMERICA: 1969-1993 Mats Lundahl * Over the course of the past twenty-five years, Swedish social scientists have cultivated a keen interest in Latin America. This is particularly true of Swedish economists. The generation born in the 1940s has devoted considerable attention to Latin American issues since the early and mid-1970s. So have a number of Latin American scholars living and working in Sweden since the establishment of military rule in Chile,Argentina and Uruguay around that same time. Since then a steady flow of works has seen the light of day. However, no attempt has ever been made to survey this particular body of literature. Hopefully, the present survey will serve to bridge that gap.
I have chosen to employ a 'broad' definition of 'economic' research. This is only natural in the present context, in view of the fact that LatinAmerica is a continent of developing nations, and in development economics in general the dividing lines between' economics', 'economic history' and' economic geography' are rather blurred. Interdisciplinary approaches are much more common here than in other fields of economic science. This implies no particular value judgment on my part. Whatever one may think, it is a reality that should be reflected in the survey. Business administration studies, on the other hand, have been interpreted as falling outside the scope of the present survey,l and so have, with a single exception, studies carried out within the disciplines of political science, social anthropology and sociology.2 I have also attempted to exclude works which obviously overlap as well as mere translations of existing pUblications.
The works surveyed herein have been grouped into the following, somewhat arbitrary, categories, which, with one or two exceptions, correspond more or less to the headings one normally finds in textbooks on development economics: To a ·certain extent, the decision as to what should be included in each category has been subjective, notably in cases where a study touches on more than one area. Basically, I have endeavored to find out where the main emphasis of a study lies and to include it under the pertinent heading, except in the case of longer, multifaceted works which may appear in more than one category.

Overviews and General Studies
In 1987, a major study by Rigoberto Garcia, Fernando Cordero andAlberto Izquierdo (Albert Patrickson) was published, dealing with the geography of economic development in Latin America. 3 This book provides ample coverage of most major issues. It is not merely descriptive, but makes good use oftheoretical principles and rests on a solid historical foundation. General overviews are mixed with more detailed case studies, in three parts, covering (1) population, urbanization (with emphasis on the twentieth century) and income distribution; (2) physical aspects and natural resources, including export products, and (3) sectoral development and economic processes (manufacturing, agriculture, foreign trade, economic integration, infrastructure and transportation and, finally, planning).
The case studies deal with population in Paraguay and the Chaco, Argentina and the Pampa in a physical context, natural resources in Brazil, foreign trade in Chile, Bolivia and the process of economic integration, and infrastructure and transportation in Mexico. The book is very thorough, as far as it goes. What is lacking is mainly a discussion of economic growth, an analysis of factor markets (labor and capital) and, somewhat surprisingly, the development of manufacturing after 1930! Unfortunately, the work was slightly outdated by the time it was published. Data and references extend until the end of the 1970s, but no further. An updated version would be most welcome.
In economic (and social) history, there is a nice little introductory text by Mauricio Rojas, regrettably in Swedish, which covers the entire time span from the period prior to the European discovery of the Americas up to the present day.4 On the one hand, an effort is made to cover the broad historical context both before and after political independence. On the other, eight countries are singled out for special treatment during a given period after independence (Argentina, Brazil, Chile, Peru. Mexico, Bolivia, Colombia and Cuba). The emphasis is on post-colonial times and on 'domestic' explanations rather than on Latin America's place in the international political and economic system.
More limited geographically, but only geographically, is Magnus Morner's impressive survey of the Andean region. 5 Although this book according to the author himself focuses on the various ethnic groups in the region -conflicts, coexistence and acculturation -from prehistoric times up to around 1980, one is nevertheless left with the impression of a work which is heavily economic, for the simple reason that it is impossible to write a history of Andean social development without digging deeply into the economic fundament. (This tendency becomes more pronounced as Morner approaches the contemporary period.) Demography, migration, labor organization and urbanization played a fundamental role in shaping social relations. So did production, landownership, taxation and trade. The history of the Andean region, after the Spanish conquest, to a very large extent, is the story of its incorporation into the international economy, through trade and investment. Export booms altered production relations, and the interplay between economics and politics was instrumental in creating a society where the struggle for economic and social reform has been the hallmark of the twentieth century. Although not an economist (or even an economic historian), Morner has provided a very rich, largely economic, interpretation of the Andean past.
A contribution to the sociology of Latin American economic theory is found in the publications co-authored by economist Magnus Blomstrom and economic historian/peace researcher Bjorn Hettne on the dependencia schoo1. 6 Their two books (the second a slightly extended version ofthe first) deal with the rise and content of Latin American dependency literature, the critique against it, its spread to other parts of the Third World, its influence on practical policy making and its influence on the later development debate. By definition, a project ofthis nature is extremely ambitious. Accordingly, its authors have been obliged to narrow their focus so as to make the task manageable. Blomstrom and Hettne have concentrated on the spread of dependency theory, choosing a largely sociological approach which rests on Thomas Kuhn's notion of scientific paradigms and paradigm shifts, 7 and by and large avoid presenting their own critical evaluation, i.e. they have avoided the perspective of traditional economists. Mats Lundahl takes a comprehensive, interdisciplinary, view of agricultural stagnation in Haiti in his extensive doctoral dissertation from 1979. 8 The major problem which Lundahl explores is the cumulative interaction between population growth and soil erosion, which tends to depress per capita income over time in rural areas. This interaction is discussed with the aid of a neoclassical model. On top of the erosion problem, however, the Haitian peasant has had to face a series of predatory governments which have taxed the peasants without providing any benefits to rural areas. This attitude on the part of the rulers has been patent in virtually all areas: nutrition, health, education, credit and technology, where government action has been sorely lacking, and this has resulted in malnutrition, disease, illiteracy, high interest rates and technological stagnation, even retrogression, reinforcing the downward spiral of productivity and incomes.

Agriculture
Imperfect markets have often been pointed to as the main culprit, notably the land and credit markets, the market for domestic crops and the market for the most important peasant cash crop, coffee. However, this hardly seems to be the case. 9 If anything, markets seem less imperfect in Haiti than in other Latin American countries. International trade cannot be blamed either. The Haitian peasant would hardly be better off had the country been 'delinked' from the world economy.1O It is the combination of population growth, erosion and bad government that has put Haiti's peasants in their present dismal situation, a situation from which they have little escape other than a purely 'physical' one: emigration. 11 The peasant economy on the other side of the border in Hispaniola has been analyzed in a thesis by Rosemary Vargas-Lundius, published in 1991. 12 The point of departure for her study is the employment problem in the Dominican Republic. Vargas-Lundius argues that this problem is not one of un employment but of poverty. People are working, but in low-productive jobs which lead to low incomes. Since the Dominican economy has basically been an agrarian one until very recently, she chooses to focus on the agricultural sector and analyzes the mechanisms which create poverty in the countryside.
Her findings are similar, in part, to those of Lundahl. The commercialization chain for agricultural outputs and inputs is hardly the culprit. Enough competition takes place to prevent that. Government intervention in the formation of prices, on the other hand, may have contributed to keep peasant incomes down. Where the Dominican Republic differs from Haiti, however, is in its agrarian structure.13 Whereas Haiti does not seem to present any major imperfections, the Dominican Republic constitutes a classical case of latifundiominifundio polarization and consequent inefficiency in production, with underutilization of the land on the former units and overutilization on the latter. This, Vargas-Lundius analyzes with the aid of the theory of distorsions. 14 The problem is further aggravated by unequal access to credit for large and small farmers and wages are in turn kept down by an inflow of Haitian workers. Over time, poverty tends to increase as a result of soil erosion, exactly as in the Haitian case. The result of all this is migration to the cities and abroad of those who are enterprising enough and who can finance the transportation.
The origins and the later development of the large estate in LatinAmerica have been surveyed by Magnus Marner in a series of works. 15 These works, which span from the early colonial period up to the present century, deal with how the institutions in the land market evolved and how they related to the institutional evolution in the labor market. The lines are drawn from the mercedes de tierras to later forms of land tenure -haciendas and estanciasand various forms of more or less dependent labor are traced. Relations between landowners and peasants in different regions and production regimes are brought to the fore.
Marner has also spent much time and effort researching landownership in the Cuzco region, mainly from 1689 to 1786, and its relation to demographic developments and market conditions. 16 The period exhibited a remarkable degree of continuity with only minor changes. The population grew only slowly and the number of haciendas changed only gradually. Marner demonstrates that the term hacienda covered a wide spectrum of properties, ranging from a minority of very large ones owned by absentee landlords to a majority of smaller ones with the owners present and only a few dozen Indian worker families. The predominance of haciendas was especially pronounced in areas with a large migrant landless Indian population. Both the haciendas and the Indian villages produced a surplus which was marketed.
The data collected during the course of the Cuzco investigation has also permitted Marner to shed some new light on the causes of the Tupac Amaru rebellion in 1780. 17 Using parish level data he shows that, even though it may be difficult to escape from the causes pointed to by conventional wisdom, the rebellion was not mainly correlated with the exploitation ofmitayos (forced draft laborers), with forced purchases of goods by the Indians or with work on the haciendas. Rather, the rebellious communities were found in areas where there were fewer haciendas, fewer Indian laborers tied to the estates, a smaller non-Indian population, and higher altitudes unsuitable for commercial agriculture. Rebellion required a certain autonomy at least from the day-to-day supervision of the European rulers and this was possible only in areas of minor economic interest.
In a joint effort spanning both history of religion and economics, Jan Lundius and Mats Lundahl have investigated another rural protest movement, led by Olivorio Mateo in the San Juan Valley of the Dominican Republic between 1908 and 1922.18 This movement, which managed to attract thousands of followers before it was persecuted and Olivorio was shot by US occupation forces, had its roots in economic change. For hundreds of years, the main livelihood in the San Juan Valley had been cattle raising on communally owned land (hatos) and the main trading contacts had been with neighboring Haiti. However, after the tum of the century, when large-scale sugar production was introduced, the hatos began to be divided in order to make it possible to sell the land. As sugar production began to dominate in Azua and Barahona, the San Juan Valley started to deliver foodstuffs to these areas. As elsewhere, the land was surveyed and the large landowners began irrigating their estates. The smallholders were threatened and a process of land concentration set in. Simultane-ously, the Americans took over the Dominican customs houses and instituted border control, turning trade with Haiti into an illegal matter. The Olivorista movement constituted a conservative peasant reaction to these changes which threatened a time-honored way of life. Olivorio stood in the way of the modernization process led by large landowners and merchants and was therefore disposed of.
Surveying the land during the American occupation (1916-24) did not solve the Dominican land problem. If anything, it contributed to the concentration process. Hence, the Dominican Republic is a clear candidate for agrarian reform. However, only half-hearted attempts have been made. A similar, and more spectacular, case of non-delivery is that of Peru after the military takeover in 1968. The Peruvian land reform is the subject of a thesis by Tom Alberts published in 1983. 19 Alberts interprets the Peruvian experiment as an example of an equity-cum-growth strategy on the part of the military junta. Unfortunately, however, he is unable to find any results in terms of income distribution. The reform turned out to be beneficial only for peasants with high incomes, relatively speaking, while the majority of the poor were not touched to any significant extent, even though as much as 50 percent of the agricultural land was expropriated. Neither was the growth objective met. The agricultural growth rate declined during the 1970s, apparently as a result of government intervention in agricultural pricing and marketing and policies favoring industry over agriculture. In this sense, the land reform did not constitute any break with the past.
Another SouthAmerican country with a recent land reform is Chile. The traditional agrarian structure in the country was much the same as in the Dominican Republic or in Peru, with monopolization of the land and concomitant monopsonization of the labor market and unequal access to credits for large and small farmers. This structure prevailed until the 1960s, although, as Mats Lundahl has shown, the degree of imperfections had probably not increased at least since the early 1930s. 20 In 1962, an agrarian reform law was passed, followed by a more radical one in 1967. The application ofthese laws destroyed the latifundiominifundio system, and when the military junta took over in 1973, it continued the process of agrarian change by creating a more or less free market for land transactions. 21 The effects ofthe latter process on the household level have been investigated in a dissertation in economic history by Paulina de los Reyes. 22 Her data come from four villages in central Chile (the Seventh Region) and the study focuses on poverty, marginalization, changes in the rural labor market and survival strategies during the 1973-89 period. With the rise of a successful commercial, export-oriented agricultural sector, production conditions worsened for small farmers, seasonal work was substituted for permanent employment, and individual employment and wage contracts replaced collective bargaining. Farm incomes among the poor had to be supplemented with other income sources and social networks assumed increasing importance as cushions in the face of irregular work opportunities. Improvisation under uncertain circumstances rather than stable survival strategies characterized family behavior.
The first land reform in independent LatinAmerica took place in Haiti in 1809. The causes behind this reform, which saved Haiti from the bimodal agrarian structure of most other Latin American nations, has been analyzed by Mats LundahI.23 The plantation system, which had been restored by the first Haitian rulers, broke down as a result of pressures emanating from both the supply and the demand side. Neither capital nor labor was available in sufficient quantities to maintain the sugar plantations. Much of the former had been destroyed during the wars ofliberation between 1791 and 1803 and labor could be mobilized only with the use of force. The Haitian administrations turned out to be too weak to prevent the ex -slaves from squatting on the land. Export markets had largely been cut off during the Napoleonic wars and the first half of the nineteenth century saw a fall in the price of sugar. Once these factors had been given free rein, Haiti became a peasant nation with a relatively equal distribution of land.
Another Caribbean country which did indeed have a land reform was Cuba, in 1959. Andrew Zimbalist and Claes Brundenius have provided an account of the contributions of the post-reform agricultural sector to Cuban industrialization up to the mid-1980s. 24 These contributions covered most of the areas traditionally dealt with in studies of the role of agriculture in economic development. Domestic agriculture had supplied an increasing share of the Cuban diet, as a result of productivity increases in the sector, labor had been released that could be employed in manufacturing, the export base had been expanded in the process of diversifying food production, linkages had been established through an increasing demand for inputs, and increased rural incomes had enlarged the domestic market for industrial products.
The choice of technique in agriculture has been analyzed in the comparative study of capitalist Jamaica and socialist Cuba by Charles Edquist. 25 The setting is sugar cane harvesting from the late 1950s to the early 1980s, a period when harvesting was partly mechanized in both countries, with the Cubans going further than the Jamaicans by mechanizing not only the loading operations but also the cutting, whenever possible. Edquist's detailed analysis rests on the concept of social carriers of techniques, i.e. the social entities that choose and implement techniques in an economy. He identifies the main actors and the constraints within which they have to operate in each one of his two cases. Thereafter, the socioeconomic consequences of the choice are explored, not least for employment and work conditions, and the different attitudes of workers and trade unions in the two countries are highlighted. The implications differed considerably between Cuba, where the main harvesting problem had been the lack of hands, and labor-abundant Jamaica, where mechanization led to human suffering and social problems. Finally, the crucial role of domestic technological capacity for facilitating change is stressed and issues of technological dependence are discussed.
Lastly, in the field of agriculture, there are two studies by Lars Pira on Central America,26 both of which deal with food security and efficiency of production. From the 1970s to the mid-1980s, the area devoted to food production decreased everywhere in Central America except Guatemala. The concentration on export crops led to land concentration and inefficiency as well. Pira surveys different measures that may be employed to reactivate food crop production, especially in the peasant sector, emphasizing pricing policies, reduction of protection and the active use of Central American integration.

Migration
Migration to Latin America from other continents as well as migration out of and within Latin America itself has been surveyed by Magnus Marner. 27 Marner's work, which deals with the entire colonial and postcolonial experience, except the involuntary migration of African slaves, divides the migration history of Latin America into four periods: the colonial era with restricted migration, the early national period from around 1810 to the mid-nineteenth century when immigration was liberalized considerably, the age of mass immigration thereafter until the onset of the depression around 1930, and, finally, the period from 1930 until the beginning of the 1980s.
Emphasis is on sketching the broad trends, providing quantitative estimates whenever possible. In a work of relatively small scale Marner is able to tell a fascinating story ofthe changing fate of migration and migrants over the course of five centuries. The currents of migration are mapped and the ethnic and socioeconomic composition ofthe migrants is discussed. Immigration policies, both during the colonial period and in the post-independence era, are given considerable weight, being crucial determinants of the magnitude of the population movements. A number of causal factors are discussed, most importantly economic ones, but other factors, including religion, politics and culture, are also brought into the picture. The impact of migration on the host countries is discussed both in quantitative terms and with respect to what the immigrants brought in terms of skills, entrepreneurship, etc., and this, in tum, is brought to bear on the role played by the immigrants in the economies.
Emigration from Haiti to neighboring countries has been studied by  Population pressure began to push poor Haitians out of their country, into Cuba, precisely at a time when the Cuban sugar industry was entering a rapid expansion phase and hands were lacking as a result of warfare in the 1890s. This migration, which took place mainly between 1913 and 1934, died out almost entirely after that period, except for a minor, politically-motivated, current following the rise of Fidel Castro to power in 1959.
Migration from Haiti to the Dominican Republic has been more stable, naturally enough, since the two countries share the territory of Hispaniola. This migration dates back to the colonial period, when the slaves on the French side discovered that treatment was better in the east and chances for manumission were relatively good. The movement from west to east continued during the period of Haitian occupation of the eastern part, 1822-44, and once sugar production expanded in the Dominican Republic at the beginning of the present century, a regular migration current could be observed. With minor interruptions, this current, which is mainly temporary, has continued up to the present, as Haitians continue to underbid Dominicans for work in the cane fields.
As for internal migration, two major studies have been undertaken during the past twenty-five years. The first one, a thesis by geographer Raul Nino Guerrero from 1975, deals with Colombia, mainly 1951 This thorough empirical study takes as its point of departure precisely the latifundio-minifundio polarization in the Colombian countryside. Nino Guerrero demonstrates how the difficulties of obtaining an income which suffices for a family triggers outrnigration, above all from those areas where most minifundios are found. This migration is age-but not gender-specific, with members of both sexes of working age emigrating, leaving the very young and the very old behind. The flows converge on two areas: the Andean nucleus, with Bogota, Cali and Medellin, and the Caribbean coast, with Barranquilla as the main center of attraction. Nino Guerrero also identifies a (quantitatively less important) movement to agricultural frontier areas. The emphasis of his study is on push factors, and he stresses the problem which the migrants face finding employment in urban areas, where they ultimately end up in the slums. To a very large extent, the migration process converts a problem of low-productivity employment in the countryside into one of open unemployment or low-productivity employment in the cities. Thus, Nino Guerrero argues that overconcentration of population results in urban areas.
The second major study, by Gunnar Malmberg, also a doctoral dissertation in geography, from 1988, deals with the growth of Lima and rural-urban migration in Peru. 3D This study, which is conducted both at the macro and the micro (interviews with migrants) levels, provides tests of different hypotheses about the causes and effects of migration. Migration stands out as a rational choice by the individuals involved, and gaps between urban and rural areas in terms of expected earnings, for example, are important determinants of migration. The Peruvian case fits the Harris-Todaro modePl fairly well. Also, urban pull factors appear to be stronger than rural push factors. Even though the migrants as a rule end up in the informal sector, where the absolute standard of living is low, they are better off there than in their rural area of origin. The informal sector solves more problems than it creates for the migrants and it is therefore possible to talk about a metropolitan informal sector pull. Malmberg refutes the idea of overconcentration in the Peruvian case. Finally, he shows that migration does not act as a balancing force between town and country. Rather, together with other variables, it increases regional imbalances in a cumulative fashion it la Gunnar MyrdaI.32

Mining and Energy
Comparatively little has been written on mining and energy.33 However, Claes Brundenius has analyzed the activities of multinational mining corporations in Peru, mainly during the postwar period. 34 The emphasis is on ownership structure, capital movements, including profit repatriation, and accumulation in a Marxist framework. Brundenius identifies a number of devices which make it possible to transfer profits covertly out of Peru: illegal exports of by-products, overinvoicing of imports and underinvoicing of exports and charging of technological fees by the mother company in transactions with its Peruvian subsidiary.35 Thomas Sterner has worked on energy policies in Latin America. 36 He has demonstrated that oil pricing policies in Latin American countries differ from those commonly pursued within the OECD. Whereas the latter group of countries tends to follow the world market price level, the former often makes active use of energy pricing as a policy instrument. State oil companies were set up in a number of oil-producing countries following the nationalization of foreign-owned enterprises, and these companies have frequently sold petroleum products at subsidized rates to industry. However, products used by poorer income groups have been sold at low prices.
Sterner has also shown that the state oil enterprises in Latin America have changed their policies drastically over time. The majority of them emerged as a challenge to the large international companies and were employed as an expression of nationalism in many countries. This, to a large extent, explains their initial emphasis on low prices for purposes of industrialization and income distribution. However, as the costs in the form of misallocation, contamination and illegal trading made themselves felt, many companies were forced to cut down on subsidization in the 1980s. At the same time, their relations with the multinational oil companies have been normalized, to the point where joint ventures, at least in individual projects, have become a real possibility.

Manufacturing Industry
The first study of manufacturing in Latin America during the period, and for several years the only one, is the doctoral dissertation from 1975 by economic geographer Weine Karlsson on the development and location of the industrial sector in Venezuela. 37 Karlsson investigates why this sector evolved relatively late, as compared to other Latin American countries. The oil sector remained an enclave which tended to generate linkage effects mainly outside the country (the Netherlands Antilles) up to 1950 and the question is asked whether a more interventionist policy would not have been better. The growth of industries producing for the domestic market (cotton textiles and three subsectors of food and beverages) is investigated. Finally, Karlsson also examines the spatial concentration to Caracas and the north-central region and the backwash effects and regional imbalances that resulted.
Industrial location has also been dealt with in some works by Albert Patrickson (Alberto Izquierdo V 8 The fertilizer industry in general was a late developer in Latin America (in spite of the availability of natural guano and nitrate), but growth in the 1970s and 1980s was rapid. Patrickson analyzes the location strategies of the Mexican and Chilean fertilizer companies during this period, and the industrialization patterns behind them, using the notions of backward and forward integration on the one hand and deliberate (precise intentions are formulated and implemented as intended) and emergent (no precise intentions) strategies on the other. Ten years elapsed between the publication of Karlsson's thesis and the next doctoral dissertation on industrial issues. This was Thomas Sterner's work on the use of energy in the Mexican manufacturing sector, both in general and in the special case of the cement industry (the most energy-intensive sector).39 The point of departure for Sterner's applied industrial economics study is the observation that during the 1970s the use of energy increased relative to output in Mexican manufacturing in general, and the question that he seeks to answer is why. The explanation is not to be found in changes in the output mix or in any bias due to technological change or scale, but in the falling price of energy relative to other factors of production. 40 The cement study concentrates on structure, technological change and substitution possibilities at the industry level.
The same year (1985), Claes Brundenius published an article on industrial development strategies in Nicaragua. 41 The article surveys the situation in the manufacturing sector when the Sandinistas took over in 1979, the changes in the following years and the industrial strategy of the revolutionary government, in the context of the collapse of the Central American Common Market (CACM) and the experienced vulnerability of foreign trade (the dependence on capital goods imports from the US).
Brundenius has also worked on capital goods production in Cuba. 42 Such production hardly existed before the revolution and during the 1960s it was hoped that a capital goods sector could rapidly be developed. However, Brundenius shows, this was a much more difficult affair than the Cubans had expected. The 1960s stand out as a decade of trial and (mainly) error, and it was not until the 1970s that the foundations were laid for what, in the mid-1980s, appeared as a sector with good prospects based, above all, on a well-educated and skilled labor force.
The most common industrialization policy in Latin America during the post World War II period has been import substitution behind tariff walls. The contradictions and tensions of this policy have been examined by Kalki Glauser for the cases of Brazil and Chile. 43 This policy rested on tariff protection of manufacturing, i.e. turning relative prices in favor of production for the domestic market. At the same time, the import-substituting industries were dependent on the foreign exchange generated by the punished sectors, notably agriculture and mining for imported inputs, and this, over time, made it difficult to continue the expansion process. Finally, the strategy broke down in both countries, but not until military governments had been substituted for the power constellations generated by the import substitution process: Brazil, 1964 andChile, 1973. Two major works on the history of industrialization in Latin America have been produced in Sweden, both by Latin Americans. 44 The first one is Rigoberto Garcia's thesis on incipient industrialization in Chile during the 1845-79 period. 45 The significance of this book is that it pushes the knowledge of Chilean industrialization back from the time ofthe War ofthe Pacific to 1845, a time when Chile had already achieved a degree of political stability and export growth. In this context, flour mills and copper foundries based on exports began to be established. Industrialization, however, did not succeed during the period studied by Garcia. (No more than 1.5 percent of the Chilean labor force was emp loyed in manufacturing in 1875.) The production of handicrafts failed to trigger a process of protoindustrialization, with cottage industries gradually being turned into modem manufacturing establishments. However, the Chilean state failed to create any fiscal linkages a la Albert Hirschman,46 spending its tax revenue unwisely and inconsistently, instead of on the development of its incipient industrial sector.
Garcia has also devoted a lengthy, critical, essay to the Chilean manufacturing sector during the Pinochet regime. 47 His article covers most aspects: changes in industrial policy (liberalization, reduction of protection, export promotion), output and employment, investment, technology, productivity, ownership, foreign trade (exports and imports) and foreign investment.
The second historical study of industrialization is that of Luis Bertola on Uruguay, 1913-61, also a doctoral dissertation. 48 Bertola provides new data and proposes a new division of the industrial sector (export-led cattle-breedingbased, domestic-market-oriented land-based, building materials, domestic-market -oriented processing of imported raw materials, state-owned oil refinement). This serves as the basis for the construction oftimes series and their subdivision into periods with fast, slow, no or negative growth. A hypothesis (which is not tested, however) is advanced with respect to the stagnation and/or decline experienced for example during the two world wars, the 1930s and from the late 1950s onward: the lack of a domestic capital goods sector creates a dependence on international markets and acts as an obstacle to technological change.
Finally, ownership issues have been dealt with by Bo S6dersten and Thomas Sterner. 49 S6dersten explores the principles governing the propiedad social sector introduced by the Velasco regime in Peru and their probable efficiency implications. Sterner, in turn, analyzes the influence of ownershipmultinationals, domestic firms, cooperatives -on the choice of technology and efficiency in the Mexican cement industry. He employs a frontier production function to arrive at measures of technical efficiency for the individual firms and relates these findings to ownership. The result is somewhat surprising in that multinationals turn out not to be more efficient and cooperatives less efficient than regular firms. It may, however, be explained on the one hand by the fact that multinationals have goals other than production efficiency when setting up affiliates in developing countries, and on the other by the inherently higher worker motivation in cooperatives.

Labor Systems
One of the time-honored areas of economic research on Latin America is that oflabor mobilization. The theme, however, is not well-developed in Sweden, with two or three exceptions. Roland Anrup has worked on the hacienda system in Colombia. 50 The issue is mainly the choice between a temporary and a permanent labor force andAnrup employs a property rights approach. Central to his analysis is the concept of disposition. The rights exercised over production factors are seen as a strategy. Through potential disposition -the capacity to expand one's own rights and limit others' -a pattern of authority is established, and through operational disposition, i.e. the capacity to determine the mode of operation of the factors, the organization of the work force is determined.
This principle Anrup brings to bear on the problem of how to establish a stable and productive labor force on a coffee cultivating estate in central Colombia since the beginning of the present century. The degree of potential disposition, which in turn by and large determined the operational disposition, has rested on the monopolization of market contacts by the hacendados. In the 1920s workers were tied to the estates by the provision of subsistence plots, but once political conditions became unstable in the 1930s, landowners gradually switched to more direct control methods, assigning particular sections of the hacienda to each worker, who then received a weekly wage to be cleared against produce delivery at harvest time.
Mats Lundahl has analyzed a similar problem in the case ofHaiti.5! He investigates how changes in the relative supply of labor and land have determined agrarian property rights in Haiti, including those in human beings. As the extent of population pressure on the land has changed, so has the property rights system. During the colonial period a plantation system was built up which, for technical production reasons, required large concentrations of both labor and capital on the sugar estates. Slavery ensured that property rights in men were strong. Once the slaves had made their revolution, however, this system was bound to collapse, since free land was available. Haiti became a peasant nation during the course of the nineteenth century.
As the population has continued to grow, access to land has become a problem in Haiti. One of the devices employed to solve this problem is sharecropping. Formerly, when the demographic pressure was not so high, a father would normally allow his children the use of smaller plots in the form of a preinheritance grant and the children would put in some labor on their parents' land. Today, the younger generation frequently become sharecroppers on the land of the older generation instead. The size of the pre-inheritance grant has been diminished and the time for it has been delayed as population growth has conferred a stronger bargaining position on the older generation and a weaker one on the younger.
The importance of demographic pressure for the choice of labor system has also been observed by Magnus M6rner in the case of slavery. 52 As Spanish America began to recuperate after the initial demographic catastrophe, a free labor force of mixed racial origin gradually became an alternative to forced labor by Indians andAfricans. M6rner also expands on Lundahl's Haitian theme and offers a condensed history of the impact of emancipation in a number of slave societies in Latin America: Barbados, Jamaica, Guadeloupe, Martinique, Colombia, Venezuela, Peru, Cuba, Puerto Rico, Brazil. 53 His perspective is one hundred years after abolition and the question he poses is what emancipation accomplished in social and economic terms for the ex-slaves when viewed over the long run. The determinants of this included demographic conditions, such as the numerical relationships between slaves, free colored and whites towards the end of the slavery period, social aspects, like the aristocratic, hierarchical character of Spanish and Portuguese American society, the prevailing production structure, the availability of land and the political system. Regional and local factors made for a great deal of geographic and socioeconomic variability and M6rner warns against premature attempts at generalization. 54 M6rner has also surveyed the writings on the 'buy or breed' question: Was imports of slaves a more cost-effective way of guaranteeing the supply of labor than ensuring high rates of natural reproduction in the plantation societies in the New World?55 He discusses the various factors shaping the decision: possible preferences for males, the supply of males and females in Africa, general economic conditions in the plantation societies, export product prices and slave prices, natural slave population growth rates, factors determining slave mortality and fertility, manumission, etc. Again, he cautions against generalizations based on incomplete material.

Foreign Trade
Foreign trade between the Old and the New World was hardly opened in the 'textbook' way. This observation constitutes the point of departure for Mats Lundahl's study of conquest and trade during the first century after the European discovery of America, which also ties in with the labor mobilization theme. 56 Conquest and trade cannot be separated in the Latin American case. The fiscal crisis in Spain during the fifteenth and sixteenth centuries was a driving force in the process of exploration and conquest. The Crown was in desperate need of revenue. The conquistadores as well were driven by pecuniary motives. However, religious factors must also be brought into the picture: spreading the Christian faith was paramount for the Church. Lundahl argues that the, partly zerosum, game between these three actors was what determined the new set-up of institutions, largely designed to ensure that forced labor was forthcoming, without which a steady flow of exports to Spain could not be maintained.
Lundahl has also dealt with foreign trade in a historical perspective in two specific Latin American countries: Chile and Costa Rica. In both cases the staples theory of growth is applied. International trade in nitrate and copper acted as an engine of growth from around 1850 to 1930 in Chile, but it is more doubtful that the fruits ofthis growth spread to all segments of the population. 57 Land concentration prevented landless agricultural workers from sharing the benefits. During the period that ensued, the import substitution policy had largely negative effects on the economy: employment, inflation, exports, food production and income distribution.
Foreign trade in Chile from 1968 to 1986 has been investigated by Kalki Glauser. 58 The bulk of his research is made up of statistical data on import capacity, import composition, exports, terms of trade and export composition. Over this period, the traditionally strong role of copper exports was weakened as exports were diversified, Brazil and Japan gained increased importance as sellers of imported goods to Chile, and exports became more geographically diversified as well.
The Costa Rican staples case differs slightly from the Chilean one. 59 Costa Rica has by and large remained dependent on exports up to the present time. Coffee and bananas appeared on the scene during the nineteenth century and quickly came to dominate Costa Rica's foreign trade. Some linkage effects were created by the new products, but the demand structure generated was concentrated on imported goods. The depression never managed to change trade policies in Costa Rica as it did in most other LatinAmerican countries. Agricultural exports, somewhat more diversified, continued to dominate in the 1950s, and it was not until the creation of the Central American Common Market in the following decade that the manufacturing industry was stimulated. This market broke down after the Soccer War between Honduras and EI Salvador in 1969, and after a difficult period of adjustment in the 1980s, the country opted for a foreign trade policy based on non-traditional products, i.e. the dependence on international markets remains, and the new strategy is not without problems. 60 The difficulties of creating new export products are also borne out by a study by Jaime Behar on Uruguay in the mid-1970s. 61 After the first oil price shock in 1973, which led to a drastic fall in the terms of trade and acute balance of trade problems, Uruguay attempted to diversify its exports away from the traditional products of agriculture and animal husbandry. The exchange rate was adjusted, and non-traditional exports were stimulated through a variety of direct measures. The share ofthese products in total exports increased from 27 to 57 percent between 1973 and 1977, but at a decreasing rate, and, as Behar demonstrates, most of the expansion derived from a few products, notably leather goods and agricultural crops (mainly rice), i.e. Uruguay by and large did not manage to break away from primary sector dependence.
Patrik We stander and Mario Zejan have analyzed the impact of further trade liberalization in Uruguay, in relation to the integration process in the Southern Cone since the 1970s. 62 Uruguay's exports to neighboring Argentina and Brazil have increased as the degree of economic integration has increased in the area, while the effects on exports to countries outside the area have been minimal. Contrary to all expectations, variations in the degree of economic activities in Brazil and Argentina do not seem to have any major consequences for Uruguay's exports.
The staples approach to foreign trade has also been employed by Lars Pira in two studies on Guatemala. 63 In the first, Pira demonstrates that the opening of international trade in Guatemala can be explained by combining the vent for surplus and staples approaches, but that these approaches must be combined with an analysis of institutional change if the variations in employment of land, capital and labor over time are to be understood. The relative backwardness of Guatemala can be derived to a large extent from its Spanish institutional heritage, not least the need for government revenue pointed to by Lundahl. From that perspective, Guatemala was relatively uninteresting and the country thus came to depend on nearby markets rather than on distant Spain.
One of the main problems of Guatemala has been the instability of export incomes and its possible relation to GDP instability. Pira has looked at two periods: 1870-1955 and 1955-85. During the former, Guatemala was extremely dependent on coffee and bananas, while the latter period is characterized by diversification of primary exports as well as sales of manufactures. He finds that instability of export prices is often neutralized by changes in the volumes exported and that the relation between changes in exports and changes in GDP is not clear-cut. The sectors producing for the domestic market must also be brought into the picture.
The importance of changes in export prices and the terms of trade for GDP growth has also been investigated by Tarmo Haavisto and Claudio Evald Nalin for all the Central American countries. 64 The economic growth of Central America is compared to that of the Nordic countries and it is shown that both groups of nations have experienced comparable variations in their terms of trade over the 1920-84 period. Thus, it is hardly these variations that explain the relatively poor performance of Central America in growth terms. The reasons must be sought elsewhere.
Haiti is another country that has relied heavily on primary exports. The pattern and prospects of these exports has been studied by Yves Bourdet and Mats Lundahl for the period from the beginning ofthe 1950s to the mid-1980s. 65 Explanations of the development of the main individual products (coffee, cocoa, sugar, sisal, essential oils and bauxite) are provided and it is concluded that the future for some ofthese products is not very bright, due to weak demand in the international market in combination with domestic quality problems. In addition, the steady growth of the population tends to drive the peasants into food crop production instead of production of export crops.
Finally, primary exports from Brazil constitute the topic of a paper by Per Lundborg.66 With the aid of a computable general equilibrium model for 1960-68, a number of trade policy changes and their effects on the distribution of incomes in Brazil are simulated: increased grain imports and decreased subsidies to domestic producers in the EEC and the United States and increased subsidies to wheat and coffee producers in Brazil. None of the foreign policy changes tends to have any effect on income distribution. Mainly the richer groups would benefit. The effects of the Brazilian policy changes, in tum, were ambiguous in the wheat case but equalizing in the case of coffee.
Trade in industrial products has been examined for the case of Mexico in a doctoral dissertation by Jaime Behar. 67 Behar conducts the analysis both on the national level and on the regional level of Nuevo Leon. The increasing role of industrial products in total exports is pointed out and it is demonstrated that variations in traditional manufacturing exports were largely a function of price fluctuations, while, in the case of non-traditional exports, instability derived from supply changes. Behar shows that manufacturing exports were positively related both to the level of effective protection and to changes in production capacity but negatively related to changes in the level of domestic demand, and offers the heterodox hypothesis (along lines originally developed by Staffan Burenstam Linder)68 that exports tend to be a residual after domestic demand has been satisfied, that the expansion of domestic demand determines the rate oftechnological progress which in tum affects supply, and that protection may be a way of keeping domestic demand up. Mexican exports were found to be relatively capital-intensive, and employment growth in them was slow, in contradiction to what standard theory would predict. Behar attributes this finding to the fact that industrial production creates knowledge and that by industrializing Mexico has been 'trading up' the ladder of comparative advantage towards products that are more advanced technologically and which require a comparatively lower labor input. On the regional level, manufacturing exports played a subordinate role. While more than half the industrial production was sold outside Nuevo Leon, less than 5 percent made its way out of Mexico . Thus, the employment-generating capacity of manufacturing exports, measured by constructing an input-output model for the state of Nuevo Leon,69 tended to be low during the period studied by Behar.
Behar has also worked on economic integration. An article from 1980 is highly polemical and attacks what Behar terms three basic hypotheses about the integration process: (1) that the actual economic system of most Latin American nations ('dependent capitalism') is the only vehicle for integration, development and regional unification; (2) that within the capitalist system, integration appears as an objective necessity, and (3) that integration is adeus ex machina / that will save us from crisis, stagnation and inequality.70 Instead, Behar recommends an approach based on sectoral and class-based interests.
Somewhat more analytical is a booklet about integration in Central America which, against the background of the prevailing economic situation in the region at the end of the 1980s, the debt problem and the problems of the export sectors, as well as the structural adjustment effort, discusses regional integration in the area: the present crisis, the prospects for reactivation of the Central American Common Market and the recent promotion of non-traditional exports. 7 ! Integration in Central America has also been examined by Rigoberto Garcia.72 Garcia finds his point of departure in two strategies for integration: the ECLA variety which sees integration as a way for the Central American countries to escape their periphery role as primary producers and industrialize instead, and the American one, intent on opening the doors to US investments, containing communism and backing dominant political groups in the region. Against this background he examines the achievements and failures of the different sectors of the Central American economies and discusses what can be done to inject new vitality into the Central American Common Market.
Finally, Jaime Behar has analyzed the effects of economic integration in the Southern Cone, viz. trade liberalization between Argentina and Brazil, following the establishment of the Programa de Integracion and Cooperacion Economica PICEAB.73 He demonstrates that indiscriminate liberalization would damage Argentine firms, that Brazil is the more technologically advanced, which tends to give the country an advantage in terms of capital goods production, and that Argentine companies producing consumer and intermediate goods would have problems attaining the minimum optimum scale level when facing Brazilian competition.

Direct Investment and Technology Transfer
One of the first major works on Latin American economic history to appear during the 1970s was Harald Runblom's monograph on Swedish companies in Latin America from the beginning of the present century to 1940. 74 In this work of case studies of AGA, L.M. Ericsson, Svenska Tandsticks AB and Bofors, based on company archives, the emphasis is on steps involved in the actual process of establishment, the nature of the companies established and the stages of establishment in relation to the activities of the parent companies. Runblom pays special attention to the companies' judgment of the political and economic climate and their (legal and illegal) interaction with the host country governments which also provides him with some general insight into the mode of operation of the latter, in particular the extent of corruption. Claes Brundenius, in his studies of Swedish companies in Brazil (L.M. Ericsson, Saab-Scania and ASEA), works much in the same tradition as Runblom. 75 The strategies employed are viewed against the background of the general economic and political development of the country. The focus is on competitive strategies and on the motives (markets, wages, raw materials) and effects of Swedish investment.
Nordic direct investments in Latin America during the 1965-78 period (1965-83 for Norway and Finland) have been covered in a study undertaken by four economists (Magnus Blomstrom, Eduardo Giorgi, Ruben Tansini and Mario Zejan). 76 Their work, which is mainly descriptive, presents statistical data for mainly Swedish, and to a lesser extent Danish, Norwegian and Finnish investments, with respect to distribution across countries, branches, employment, capital, activities, etc.
The above studies concentrate either on individual companies and their histories or the empirical picture at the aggregate level. Peter Svedberg's doctoral dissertation on foreign investment in Latin America, from 1977,77 has a completely different perspective. There, the emphasis is on the development of general economic principles and Latin America comes into the picture only as an illustration of these principles. The issue is the policy of a small country visa-vis multinational, monopolistic companies working under conditions of decreasing costs. Should the country in question put tariffs on imports from these companies and should it encourage the establishment of subsidiary companies inside the country? Which policy instruments should be used to bring about this establishment?
Svedberg demonstrates, in a partial equilibrium setting, that an optimal tariff exists which maximizes the difference between the tariff revenue and the loss of consumer surplus that arises from the tariff and that this tariff can be positive. He also discusses the notion of a switchover tariff, i.e. the tariff that establishes subsidiary production inside the country, compares this tariff with the optimal one, and demonstrates that when establishment of subsidiaries is desirable, from the point of view of the host country, subsidies are to be preferred to tariffs, since subsidies cause fewer distortions of the allocation of resources. The theoretical results are thereafter applied to a discussion of American multinationals in Latin America: the pros and cons of subsidiaries as compared to imports, including the balance of payments and terms-of-trade effects and the spread of technology and knowledge via foreign subsidiaries.
The shift of emphasis towards employing LatinAmerican data strictly for purposes of illustration or for tests of economic hypotheses in general is even more pronounced in the work by Magnus Blomstrom on the spillover efficiency of foreign investment in Mexico. 78 (This country has been chosen mainly because it is one of the few developing countries that have reasonably good statistics on the activities of foreign-owned affiliates.) Blomstrom employs data mainly from Mexican censuses of manufactures for 1965,1970 and 1975 to investigate to what extent the presence of foreign multinationals in Mexico results in in-creased productivity also in domestic firms. The performance of foreign and domestic firms are compared in terms of labor productivity, capital intensity, wage level, profitability, etc. Differences in technical efficiency among Mexican plants are related to the presence of spillover effects and possible mechanisms of transmission are discussed. Finally, the effects of the presence of foreign multinationals on the degree of concentration of production in different sectors is discussed.
Mexican data have been employed also by Ari Kokko in three chapters of his doctoral dissertation on host country characteristics and spillovers.79 One of these analyzes industry-level determinants of technology imports by foreign Mexican firms in 1975 and focuses on the importance of competition from local firms and the skill level of the labor force for technology acquisition by affiliates of foreign multinational corporations. Another examines the technology gap between affiliates and local firms with the aid of 1970 data in a number of manufacturing subsectors and demonstrates that technology characteristics have little to contribute to the explanation of spillovers. The latter occur both in industries characterized by large gaps and in branches where the gaps are small. Only when markets are simultaneously characterized by large foreign shares and a large technology gap do spillovers tend to be absent. Finally, Kokko also estimates the labor productivity in foreign and local firms to highlight the influence of competition on spillovers.
Like much other work in this field, the spillover studies to some extent have the character of 'measurement with ad hoc theorizing', for lack of a welldeveloped consistent theoretical framework. This is also evident in Blomstrom's works on the role of multinationals in LatinAmerican exports,80 where some of the main findings are the following: (1) US and Swedish multinationals appear to differ in their behavior, with US firms using their affiliates as platforms for exports while Swedish companies aim for the domestic markets of their host countries; (2) US affiliates in the service industry are present to a much lesser extent than in the goods-producing sectors and do not differ much from their parent companies or affiliates in developed countries in terms of physical or human capital intensity, presumably as a result of the difficulty of arriving at a global division oflabor within the multinational framework. Services are much harder to export than goods, and (3) US multinational affiliates tend to be more flexible than domestic firms and hence better equipped to make the transition from production for import substitution to production for the export market.
The direct transfer of industrial technology from one country (Sweden) to another (Uruguay) has been studied by Ruben Tansini in the context of the dairy industry.8l (Uruguay buys most of its dairy equipment from Sweden). An explicit comparison is made between milk reception and milk treatment processes in the two countries. Average and frontier production functions are estimated. The main findings are that Uruguay has little choice other than to buy the latest available technology, which happens to be highly labor-saving. However, some adaptations to local conditions can be made, mainly by using more labor and less capital in auxiliary activities which makes it possible to offset the technological bias to some extent but which also makes for a slightly lower average efficiency.
The role of technology policies has been emphasized in a comparative study of telecommunications in Brazil and India by Claes Brundenius and Bo G6ransson. 82 The history of Brazilian technology policies is surveyed. In the 1960s a strategy was developed that envisaged Brazilian control over the telecommunications service sector, the 'nationalization' of technology and the creation of a research and development infrastructure. Twenty years later, the control goal had been reached and Brazil possessed a telephone system in good shape, Brazilian subcontractors played an important role and a functioning R&D environment was beginning to take shape, at a substantial cost, however.

Finance and Debt
In 1982, the Institute of Latin American Studies in Stockholm delivered a report to the Ministry of Foreign Trade on the lending policies of the Inter-American Development Bank in Latin America. 83 The study concludes that in spite of increased lending during the 1970s, the share of the IDB in total lending to the region declined. Still, the IDB played an important role for smaller and poorer countries, although Brazil, Argentina and Mexico accounted for around 45 percent of the total at the end of the decade. In the main, poor groups were not directly targeted for lending and lending to the social sectors had decreased, relatively speaking, since the 1960s. Instead, expansion was related to the energy sector, in the wake of the oil crisis. By and large, non-democratic political regimes seemed to be somewhat favored over democratic ones, for unknown reasons.
The role of multilateral lending institutions in Latin American finance has also been discussed by Weine Karlsson,84 who points to the diminished role of these agencies in the 1970s, discusses the consequences of the admission into the IDB of non-American members in terms of lending policies and proposes reform measures: more influence for the borrowers, improvement ofthe position of the least developed members, new lending channels and more funds to lending agencies operating on the regional and subregional levels.
During the 1970s the foreign debt situation of most Latin American nations deteriorated dramatically. The debt crisis has been the subject ofinvestigation for a number of authors. Weine Karlsson and Stefan de Vylder have both produced overviews of how the crisis developed and how, as a result, the 1980s turned into a 'lost' decade in much of Latin America. Both authors also discuss various options for facing the crisis. 85 De Vylder has also dealt with the debt situation and its causes in Argentina, Chile and Uruguay against the background of the neo-liberal economic policies pursued. 86 Juan Alberto Fuentes and Lars Pira have analyzed the Central American debt problem. 87 They examine the developments that led to indebtedness and demonstrate that the debt situation in the region has been far from homogeneous, with Nicaragua and Costa Rica being far worse off than Guatemala and El Salvador. Debt service and renegotiation problems are also discussed and the desirability of a joint renegotiation of the Central American debts (due to weak individual bargaining positions) is highlighted.
A few studies on the situation in individual countries have also been published. Renato Aguilar, Eduardo Giorgi, Ruben Tansini and Mario Zejan have simulated different scenarios of debt repayment for Argentina and Uruguay: 'growth', 'expansion', 'depression' and 'moderate growth' . 88 It is demonstrated that in both cases domestic economic policy will not be effective in terms of influencing repayment capacity. Instead external circumstances must be changed, i.e. debt conditions should be renegotiated. Jose Gofii has worked on the Chilean debt: the external and internal reasons, the evolution of the debt, the problems of repayment, renegotiation and conversion and, finally some ofthe consequences of foreign borrowing, especially the support for the economic policies of the military govemment. 89 He has also examined the need for a coordinated policy among the Latin American countries in order to provide a strong platform in the renegotiation of the debt with the international banks. 90

Disaster Relief
Disaster relief may be thought of as a very special form of resource transfer: a transfer intended to compensate for physical and other damage resulting from natural disaster. However, as Galo Abril-Oj eda has demonstrated in a doctoral dissertation that focuses on the 1976 earthquake in Guatemala (7.5 on the Richter scale), disaster relief may have a significant development effect. 91 Although conceived of as a short-term measure, it has long-run consequences as well. 92 Abril-Ojeda's book contains a wealth of information. He estimates the physical damage at some 30 percent ofthe 1975 GNP and the damage to human capital (deaths and injuries) as having affected close to 2 percent of the total population. The earthquake also had serious distributional consequences in that it hit mainly the poor, who were simultaneously the ones segregated to the most disaster-prone areas and the ones least covered by insurance or other safety nets and who were in addition hit by adverse movements in relative prices. Finally, in the absence of relief measures, output would have fallen but, as Abril-Ojeda demonstrates, the assistance obtained from abroad actually made the growth of GDP in Guatemala increase to a level which lay above the level which would have prevailed in the absence of the earthquake for the 1976-79 period.
It is thus obvious that disaster relief was not limited to purely compensatory activities but in addition had positive spillover effects. Microeconomic evi-dence indicates that in the extraordinary situation after the catastrophe, the quality of physical capital was improved, new physical capital was added, which made it possible to increase employment, knowledge was improved, etc. People were reached who would never have been touched by 'normal' development assistance. Non-governmental organizations in particular were successful in implementing projects with long-run effects.

Economic Relations between Latin America and Sweden
As should be expected, the economic relations between Latin America and Sweden have attracted the attention of a number of researchers. 93 Magnus Marner has briefly surveyed the period from 1600 to 1945: the slow beginning of trade and Swedish emigration, above all to the Southern Cone during the nineteenth century.94 Swedish emigration to Brazil and the return migration from there between 1850 and 1940 has been explored by Garan Friborg. 95 This emigration took place in three waves: an initial, very small one, consisting of a mere 166 people, 1868-73, a second 'feverish' one, 1891-92, with a total of some 2,000 people mainly from Stockholm and Sundsvall, and a third, following the general strike in 1909, with about 700 people. These emigrants, proletarians in their majority, in the main failed to eke out an existence in the harsh border areas of southern Brazil. Around 600 returned home, with the aid of the Swedish legation in Buenos Aires, and another 500 emigrated to Misiones in Argentina.
The period from the beginning of the present century has been covered in more detail by Weine Karlsson and Ake Magnusson. 96 After the turn of the century, expanding Swedish manufacturing companies began to acquire an interest in the middle-income Latin American countries. In 1904, the first direct Swedish freight line was opened and the same year a decision was taken with respect to the first Swedish direct investment, L.M. Ericsson in Mexico. Ericsson was soon to be followed by a number of other companies. 97 The LatinAmerican market expanded rapidly during the 1920s and provided a substitute for the loss ofthe important Russian market. At the beginning ofthe 1930s, some 30 Swedish subsidiaries were operating in Latin America.
Stagnation during the depression years of the 1930s was followed by a new offensive during and after World War II. In 1945 Latin America accounted for no less than 19 percent of Swedish imports and 16 percent of our total exports went to that region,Argentina accounting for around half the trade in each case. Swedish companies intensified their establishment endeavors in the continent. This constituted the peak of Sweden's economic relations with Latin America. Thereafter, a decline set in as import substitution behind high tariff barriers became the rule, until the share of Latin America in total Swedish exports and imports had fallen to below 2 percent at the beginning of the 1990s.
Until the 1980s, Swedish companies had continued to expand their production in Latin America, but with the advent of the debt crisis, the momentum of this expansion was lost.
Swedish economic relations with Latin America have been heavily concentrated on Brazil during the last few decades. These relations have been studied by Rigoberto Garcia. 98 Imports reached a peak in relative terms at the end of World War II, while exports peaked some ten years later. Thereafter, however, Swedish trade with Brazil followed the general downward trend, with very large export swings, due to the impact of the oil and debt crises on Brazil's economy. Swedish credits to Brazil have been of minor importance while Swedish direct investments in Brazil represented almost one-third of the accumulated total Swedish investments in Latin America at the end of the 1980s, but only 2 percent of Swedish foreign investments globally.
A 'different' perspective on Swedish, and other Nordic, economic relations with LatinAmerica is taken by Magnus Blomstrom and Patricio Meller, in a summary of a comparative project. 99 The comparison attempted is a distant one, but the purpose is to explain differences in development paths between Chile, Colombia, Ecuador and Uruguay on the one hand, and Denmark, Finland, Norway and Sweden on the other. The thesis advanced is that the recurrent economic crises experienced by the Latin American nations during the present century could have been avoided if they had followed a 'Nordic' model, with land reforms, investment in education, emphasis on industrial products based on natural resources, outward-oriented trade and industry policies, reliance on international technology and avoidance of direct government intervention in production.

Transportation
A major study on transportation, by historian Heman Horna, focuses on the role of Cuban engineer and entrepreneur Francisco Cisneros in railroad building and development of river traffic in Colombia between 1874 and the mid-1890s. 100 Cisneros, a Latin American version of Andrew Carnegie, IP. Morgan or Cornelius Vanderbilt, came from an aristocratic Cuban family. He studied railroad engineering in Havana and New York and worked on narrow-gauge railroad development in Cuba before getting involved on the insurgent side in the 1868-78 war. Cisneros also worked on railroad construction in Peru before being invited to Colombia in 1873 to construct the Antioquia railroad.
The Antioquia venture constituted the beginning of twenty years of intense activity in Colombia for Cisneros on the Cauca, La Dorada, Girardot and Bolivar railroads and in steamboat transportation on the Magdalena river. In the late nineteenth century, communications were difficult in Colombia, especially in the interior, and the constructions of Cisneros made it possible to break the relative isolation and to develop the country's future staple export: coffee. Horna's fascinating account allows the reader to follow a capitalist entrepreneur working in an extremely difficult environment where competitors did not hesitate to resort to unconventional devices and political influence and where a civil war (in 1885) interfered with construction activities, his own shrewd manipulations to defend his position, his efforts and problems when it came to raising capital for his ventures, not least securing government subsidization and its actual disbursement, and securing the necessary labor force, his rise to national prominence and, finally, his withdrawal from the Colombian scene to devote his last years (Cisneros died in 1898) to the cause of Cuban emancipation from Spain. Horna's book is a little gem of entrepreneurial history.

Development Strategies and Economic Policy
Jorge Buzaglo has worked on development strategies in both Mexico and Argentina. 1Ol He employs a dynamic model based on a social accounting matrix and simulates the effects of different development strategies, defined as combinations of public investment and income distributions. In his doctoral dissertation on Mexico, the period under study extends from the mid-1970s to 1990, and Buzaglo compares the effects of the official Plan Global de Desarrollo, with emphasis on self-reliance, employment, basic needs, growth and distribution, a strategy based on investments in the oil sector, no redistribution of income and no agrarian reform, and finally, a 'new' strategy where the emphasis has been put on investments in other sectors, notably agriculture, with redistribution and institutional reform. It is demonstrated that the last of the three strategies performs better in terms of growth, distribution and welfare than the first two.
The Argentine study employs the same type of model as the Mexican one, with given prices and techniques as the main limitations. Buzaglo begins by sketching two fanciful 'scenarios' with respect to the future of the international economy, one where depression and high unemployment are combined with inflation and protectionism, and one where the world economy becomes increasingly integrated, with freer trade and a world currency. These scenarios serve as backgrounds for the policies simulated (for a ten-year period following 1985). The first two simulations assume an 'intermediate' state of the world and investigate what would happen on the one hand if Argentina had ceased payments on its foreign debt and the efficiency of investment remains low, and on the other hand what a minimization offoreign trade would lead to. In both cases the result is a growth rate close to zero. Should the world economy disintegrate, the latter policy would increase domestic industrial production and growth also in the case where substantial income redistributions are carried out. Finally, in the integrated scenario, making use offoreign trade allows Argentina to grow fast and to payoff its external debt.
Argentina's development strategies during an earlier period have been investigated in two essays by Carl Magnus Grenninger and Eurique Ganuza, respectively. 102 Grenninger deals, mainly descriptively, with the period from the 1930s to the late 1960s: the reconquest of power by the landed oligarchy after the fall ofYrigoyen in 1930, the rise of populism with Peron, and the subsequent stagnation of the economy, the efforts to establish a new order after 1955 and the problems facing Argentina in the 1960s. Ganuza's essay is more analytical. It spans the period after the fall of Peron in 1955 until the late 1970s. Within a Marxist framework, Ganuza characterizes the various economic regimes that held sway during the period and in particular the ways in which stimulation of capital accumulation was attempted in the 1970s.
One of the first countries to receive major attention during the period presently under consideration was Brazil. Three Marxist-inspired studies from 1969 and 1970 all stress the negative impact of foreign influences on the country's development strategies and the results of these. The major study is that of Tomas Gerholm, a social anthropologist, and Irene Matthis, a physician. 103 The 'economic' (and major) part of their book sketches the history of Brazil from 1500 to 1970. The creation ofthe class-based society and its change over time, under 'colonialism', 'British neo-colonialism' and' American imperialism', is discussed. Against this background, the authors attempt to demonstrate that the crisis of Brazil in the late 1960s was a crisis of capitalism. The latter part of the book enumerates the main political organizations on the left and analyzes the revolutionary possibilities that these organizations offer.
Two essays, one by Anders Sjoberg and the other by Lars Kronvall, offer a similar perspective. 104 Sjoberg covers the same period as Gerholm and Matthis and concludes that ever since the beginning of the sixteenth century the power in Brazil has been in the hands of foreigners, while the indigenous popUlation has had no influence on its fate. Since the end ofthe colonial period American and Western European interests have dominated the scene, together with their allies among the Brazilian elite. Kronvall picks up the same theme in an Andre Gunder Frank inspired 105 effort to demonstrate that Brazil has been a dependent satellite in an imperialist-dominated world.
A much later analysis of Brazilian development strategy, the doctoral dissertation by economic historian Wilson Do Nascimento Barbosa, from 1984, deals mainly with growth. 106 Barbosa compares two growth strategies -import substitution-based industrialization 1940-64 and export-oriented industrialization 1964-80 -with respect to content and growth effects. Content and policy environment are described at great length, and regression analysis is employed to establish relations between such macroeconomic variables as savings, disposable income, consumption and taxes. The results are not very clear-cut and are difficult to interpret.
Another country which has received a fair amount of attention from Swedish researchers is Chile. Stefan de Vylder has looked at economic policies from a historical perspective, beginning with the colonial period, but with emphasis on the period from the depression of the 1930s to the late 1960s. 1 0 7 The thesis advanced is that the Chilean economy has always been a failure in terms of development. Secular stagnation has been the rule and the economy never managed to 'take off'. Whatever growth has taken place in a particular sector has either failed to spread to the rest of the economy or has been of only limited duration. The failure to remedy the structural defects has naturally made for perpetuation of underdevelopment.
Mats Lundahl has analyzed the connection between political and economic developments during the 1952-73 period, from the return of Carlos Ibanez to the presidency to the fall of the Allende government. 108 He traces the rise of the Christian Democrat party with its emphasis on reforms, the gradual unification of the left, culminating in the election of Allende, and the gradual erosion of the economy on the one hand and of the political center on the other, both of which contributed to the polarization of politics which ended with the military coup in 1973.
The major work on Chilean development strategy and economic policy is Stefan de Vylder's monograph (dissertation) on the Allende episode, one ofthe first to appear internationally and still a standard work. 109 Against the background of the economic and political situation that had developed up to 1970, the economic strategy of the Unidad Popular is sketched. The short-run facetsin terms of income policy, public expenditure, monetary policy, price and supply policy -and their favorable impact during 1971 are discussed. Thereafter, attention is turned to the far less successful 1972 and 1973, with spreading shortages, accelerating inflation and stagnation as well as increased political tension between the Unidad Popular and the right. De Vylder also analyzes the attempts of the Allende government to restructure the economy through nationalizations and agrarian reform, as stepping stones on the road to a socialist society, and their results. The main theme running through the book is one of critical distance to the actions of the Unidad Popular on the economic front which are seen as a policy that was doomed to fail, because it was full of contradictions and hence simply did not work. Short-run expansionist gains were traded for long-run disaster.
A critical attitude is also apparent in Nancy Barrett's and Bo Sodersten's analysis of the Allende period, where the leading theme is the lack of a reformist perspective among the Unidad Popular and the application of a doctrinaire Marxist perspective in Chilean economic policy. 110 The failure lay in the inability to create short-run stability in the economy in a situation characterized by increasing political polarization.
A number of evaluations of the development strategy of the junta period are also available. Those written during the first ten years after the coup are generally critical about what had been achieved in terms of economic policy. Thus, in 1979, Peter Englund and Lars Heikensten pointed out that the military regime had managed to reduce the rate of inflation to the levels prevailing in the 1960s, but only at the cost of a drastic contraction of the economy. Few signs pointed towards rapid growth and development for the future. II I The same pessimism is evident in an article by Stefan de Vylder from 1982 which traces the developments up to 1981. 112 De Vylder reports about success in terms of inflation control, and failure in terms of employment and income distribution. As a result, consumption patterns were biased towards imported goods. The military regime also failed to increase the share of investment of GDP and investment tended to have a speculative, short-run bias. Private investment failed to compensate for public cutbacks and Chile did not manage to attract much foreign investment. The fact that Chile was a dictatorship could easily lead to firms being misinformed. Finally, the openness of the economy, among other things, made the latter vulnerable to shocks.
Cautious pessimism also characterizes two articles by Mats Lundahl, published in 1982 and1983.113 The first evaluates Chilean economic policy during the first ten years of military rule. The privatization drive, the transition to a market economy and the changes in stabilization policy are given the most attention and the difficulties in finding a yardstick against which to judge the outcome of all this is highlighted. The reduction of the rate of inflation and of the budget deficit and the increase in non-traditional exports (criteria of relevance for foreign donors) had to be weighed against low growth, increased inequality and high unemployment and an economic policy that after 1979 had contributed to recession. Lundahl concludes that more attention should have been paid to employment and less to bringing the rate of inflation down to 6 percent.
Lundahl's second article deals with the short run, i.e. on the stabilization policy between 1979 and 1982. At the root of this policy was the monetary approach to the balance of payments, according to which combating inflation is a task for fiscal policy, while monetary policy has no effect. This stabilization policy, which assumes flexible markets, was carried out mainly with the long run in mind: the transformation of the Chilean society and economy. In the short run, however, the current account ofthe balance of payments remained in deficit, even though it should have been in equilibrium according to the monetary approach. Simultaneously, the government refused to devalue the currency and the real rate of interest remained high, a situation conducive to recession, until a devaluation finally took place in 1982.
The last years of the military regime have been incorporated in.to the more recent works of Stefan de Vylder. 1l4 Much of his earlier criticism remains in place, but de Vylder also acknowledges the positive changes that took place after the 1982-83 crisis: a structural change which made the economy more competitive internationally, macroeconomic stability with comparatively low rates of inflation, a large surplus in foreign trade, a substantial reduction of the foreign debt, high growth and a rapid reduction of unemployment, and an inflow of capital instead of an outflow. He points out that after the crisis years, junta policy turned more pragmatic and less die-hard neo-liberal. Above all, it turned more expansionist with successive devaluations, a more active employment policy, and state intervention in the shaky banking sector, to the point where the new democratic government after the 1989 elections had to take over an economy that was considerably overheated.
The Peruvian case was subject to monographic treatment twice in the 1970s. The first time was in Claes Brundenius' 1972 book on the effects of four hundred years of imperialism in the country. I 15 Brundenius begins alreday with the pre-Incan and Incan cultures, follows 'the development of underdevelopment' during the colonial era of 'mercantile capitalism', 'industrial capitalism' with the guano and rubber booms during the late nineteenth and early twentieth century and the age of 'monopoly capitalism' extending into the contemporary period, dominated by increasing American influence in Peru. The second part of the book concentrates on underdevelopment during the post-World War II period. Growth and income distribution figures are contrasted, the important fishmeal and mining sectors undergo detailed analysis, with special emphasis on distributional aspects. 116 Foreign trade is examined in terms of unequal exchange and the industrialization process is placed under critical scrutiny. Finally, the prospects for reform -especially land reform -are discussed against the background of the guerilla movements ofthe 1960s and the takeover by the military in 1968. For the reader who can overlook or ignore some of the more pronounced Marxist parts, Brundenius' work still has a lot of insight to offer.ll7 The second monograph, from 1977, is the product of teamwork by Peter Englund, Lars Heikensten and Claes Nordstrom. 118 Their focus is on the 1968-75 period and the question raised is whether the strategy of the Peruvian generals was likely to be conducive to economic development on the one hand and whether they and their policy would be able to survive politically in an environment which was increasingly characterized by pressure from the United States and Peru's increasingly 'reactionary' neighbors. The authors focus on conflicts between the attempts to bring about long-run structural change and the need for short-run economic stabilization and analyze the problems faced by the various individual sectors ofthe economy, notably the land reform effort in agriculture and the problem of capital intensity and employment in the small manufacturing sector. Peru's external dependence (mining, fishmeal and manufacturing) is discussed and the importance of popular participation in the decision process is stressed. The book ends on a negative note, with the rightward turn within the military in 1976 after the fall of General Velasco the year before and the fear of a halt in the reform process.
The small Central American economies have received much less attention than the larger states, with some exceptions. Stefan de Vylder has written on Nicaragua's development strategy during the Sandinist regime. 1l9 An evaluation from 1982 focuses mainly on the 1979-80 emergency program and the subsequent 'reactivation' phase, 1980-82, i.e. mainly before the outbreak of Contra warfare, while a second article, from 1989, draws conclusions with respect to the entire Sandinist period. The economic strategy of the Sandinist government is described and its initial success and subsequent failure are explained, against the background ofthe war and the American trade embargo.

Magnus Blomstrom and Mats
Lundahl have surveyed the economic problems of Costa Rica mainly in the 1980s: the decreased dependence on traditional exports, industrialization and import substitution in the context of the Central American Common Market, the 'lucky' 1970s, the debt crisis and the efforts to create a new export base in the 1980s. 120 El Salvador has been dealt with in a similar fashion by Goran Holmqvist, Anders Pedersen and Torgny Svenungsson. 121 The authors briefly summarize the country's economic history during the present century, but the main focus is on the economic policy of the 1980s and the recent neo-liberal reform program. Problems of foreign trade and the dependence on aid and remittances from emigrant workers are also brought to the fore.
Cuban economic development has essentially been the territory of Claes Brundenius, although the first book on the subject was written by GunnarAdler- Karlsson in 1971.122 Adler-Karlsson examines Cuba's relative failure or success in the economic sphere after 1959 and how this is related to the development policy pursued by the Cuban government. His main thesis is that it is impossible to combine redistribution a fa cuban a with economic growth. Foregoing the latter is the price that has to be paid for the former, simply because the funds available for investment are limited so large-scale spending on the social sectors by definition excludes spending on enhanced goods production as well. This lack of capital is an 'inheritance of underdevelopment' and so is the lack of technology and skilled and specialized labor as well as the essentially capitalist mentality of a population intent on competing instead of cooperating with one another. More importantly, redistribution on the Cuban scale necessarily entails encroaching on the privileges of the clases hacenadas who at the outset are those controlling the means of production. Other factors that have made development difficult, according to Adler-Karlsson, are the sanctions imposed by the United States, the fact that socialist development policy was something untried hitherto in Latin America and hence necessarily a trial-and-error process. Finally, the egocentric personality of Fidel Castro and his emphasis on ideology has played an important role. Clearly, this analysis still has something to offer some twenty-five years later.
Brundenius' theme is much the same as that of Adler-Karlsson: what has the Cuban revolution accomplished and where has it failed? An extended and revised version of his doctoral dissertation concentrates on the problem of arriving at meaningful statistical series that can serve as base material for an evaluation. 123 Brundenius presents time series for national income aggregates from the beginning of the present century up to 1958. For the 1959-81 period, using official figures of the total material product as his point of departure, he calculates GDP at factor cost in constant prices. The 1960s stand out as a decade of few changes in the economy, whereas the following decade witnessed rapid growth in the industrial sector and of capital stock. Brundenius also calculates an index of basic needs performance, including food and beverages consumption, clothing, housing, education and health, which shows that while not very much was accomplished during the 1960s, the 1970s witnessed rapid growth of basic needs fulfilment. Finally, the changes in the distribution of income are calculated, an exercise which indicates a considerable reduction of inequality up to 1978.
Brundenius' second major work on Cuba is a book co-authored with Andrew Zimbalist. 124 Here, the perspective is widened. Again, statistical problems are discussed. 125 The official figures for industrial production are recalculated and GDP figures are estimated both in Cuban prices and in dollar prices. The results are comparable with those ofBrundenius' earlier study: slow growth in the 1960s but rapid growth from 1970-85. A case study of the rapidly growing capital goods industry is presented the importance ofthis sector for technology and learning is stressed. Another key sector in the Cuban economy is agriculture. Zimbalist and Brundenius show that considerable diversification -away from sugar -took place, and that within the sugar sector itself by-products have come to playa more important role. Still, the subsidized price for sugar paid by the Soviet Union committed the Cubans to continued expansion of sugar output. The awkward management and incentive system prevailing in the centrallyplanned Cuban economy and the concomitant inefficiency problems are discussed and decentralization measures are recommended. The foreign trade situation, in particular the extent of dependence on sugar exports and the Soviet Union, as well as of the extent of dependence on Soviet aid is brought up and the authors claim that in both cases the generally prevailing view of extreme dependence is exaggerated. Finally, the Cuban performance is put into an international perspective and Zimbalist and Brundenius conclude that both in terms of growth and equity the Cuban economy has performed far better than most developing economies.
Development strategies in Jamaica are the subject of a paper by Joran Carlsson who traces all the swings of the pendulum from the 'development by invitation' strategy during the latter part of the period and the 1960s, via the trade union dominated radicalism under Michael Manley in the 1970s, to the rightward orientation during the government of Edward Seaga in the following decade and, finally, the much less radical approach of the recent Manley government, which represents less of a break with the Seaga policies than expected.
Geographer Irma Guillen has evaluated a regional development project in Venezuela. 126 In 1960, the Venezuelan government started to establish a 'growth pole' in the isolated southeastern part of the country to develop the important natural resources of the region. Ciudad Guayana, which was later (1980) to have more than 300,000 inhabitants, was created, through a new organization, the Corporacion Venezolana de Guayana. Steel and aluminium industries and electric energy plants were set up during the first twenty years of the project's existence.
Guillen examines the distributional implications of the project up to 1980 and demonstrates that the development of Ciudad Guayana led to a process of social and geographical segregation between the city's two main parts, Puerto Ordaz and San Felix, not least with respect to housing and other urban services.
The inequalities in terms of incomes and living conditions that are typical of other Venezuelan and LatinAmerican cities were also reproduced in the case of Ciudad Guayana.

Inflation
One of the 'classical' themes in Latin American economics is that of inflation. This has been the subject of three major studies and one minor one in Sweden. The first major one, a dissertation by Jorge Dresdner on Chile, deals with what has perhaps been the major controversy: structuralist versus monetarist explanations of the causes of price increases. 127 Dresdner constructs a general equilibrium model which allows him to compare the two schools. He spells out the crucial features of structuralist inflation theory: stagnant food supply, downward rigidity of non-food prices, downward rigidity of wages and a monetary policy that accomodates structurally determined price increases. Of these, Dresdner concentrates on the food supply bottleneck: population and income growth leads to an increased demand for agricultural goods, but the foodproducing sector does not increase its supply, food prices rise and the rise spills over into the rest of the economy.
This hypothesis is tested formally for Chile during the 1950-72 period. Dresdner demonstrates that food production was lagging behind and discusses some possible reasons for this phenomenon. He then constructs a model which contains one 'structural pressure' block which determines the values ofthe real values, and one 'propagation mechanism' which transmits the structural inflation impulse to the nominal variables. The influence of the food bottleneck on the price level, however, turned out to be almost negligible. The lagging food hypothesis of inflation thus received no support at all. The structural impulse was small. Instead, the most important element of the inflation process appeared to be the powerful propagation mechanism which made exogenous influences on important prices in the economy have an impact on the price level.
The second major study is that of Eduardo Giorgi on UruguayYs In this study, the emphasis is on deducing reduced forms from theoretical models which can be employed to test the explanatory power of monetarist, structuralist and neostructuralist approaches to inflation. The period investigated is 1956-85 -a period with considerable changes in foreign trade and exchange rate regimes.
Two sets of models are tested, one derived essentially from closed-economy approaches and another based on models for the open economy which allow for a distinction between traded and non-traded goods. The results obtained do not allow for unconditional acceptance or rejection of any of the approaches investigated. Rather, they point to a number of regularities in the inflation process. Thus, in the 'closed-economy' formulations there was a stable relationship between the money supply and the rate of inflation. The Natural Rate of Unemployment Hypothesis also received some support. Monetary measures quickly translated into inflation while the level of activity was left unaffected. The relation between inflation and output was weak both in the short and in the long run. A disequilibrium formulation, on the other hand, suggested that effective demand changes did affect output.
Turning to the 'open-economy' approach, in the monetarist framework, inflation emanated mainly from the domestic credit market up to 1974 when the economy was opened up, while propagation of foreign price increases via the balance of payments became an important element thereafter. The shift to a flexible exchange rate regime did not impede this. Monetary policy did not become more effective. Allowing for the possibility of disequilibrium in the real sector of the economy revealed that expectations about the prices of nontraded goods could affect the rate of inflation. Estimation of structuralist models gave the price of agricultural products a role, and a neo-structuralist cost push model revealed that increasing interest rates could both feed inflation and reduce output. Thus, the conclusion reached by Giorgi is that the monetarist/ structuralist debate is far from resolved and that failing to take supply-side imperfections into account may lead to excessive contraction when orthodox stabilization measures are employed.
The third major inflation study is the thesis by Dick Durevall on Brazil. 129 Durevall poses three different questions. The first is whether in Brazil from 1968 to 1979 the adoption of a real exchange rate policy rule (indexing the exchange rate change to the difference between domestic and foreign rates of inflation) made it possible to determine the rate of inflation independently of what went on in the rest of the world. For this to be the case, changes in domestic credit should have an immediate effect on inflation and the exchange rate should then adjust to ensure balance of payments equilibrium. For the latter to be the case, the balance of payments should react faster, by changing the foreign reserve component of the money supply, which means that monetary policy would become ineffective. Durevall's tests failed to indicate any such loss of effectiveness.
Durevall's second query is whether 'money causes prices' or money should not be included among the determinants of the price level. Using monthly Brazilian data for 1968-85, he tests for Granger-noncausality to find out whether money stock variables drive the price level or vice versa. His main finding is that money cannot be excluded from the price level equation, but the choice of money variable also influences the outcome, so that when broader aggregates are employed the monetarist thesis receives support whereas using the monetary base makes a Keynesian formulation where inflation is the result of excess demand for goods and of the wage level more appropriate.
Finally, Durevall employs 1968-85 data to test for the relative importance of excess demand (money, foreign exchange, labor) and inflation inertia (past inflation) as determinants of price level changes. Inertia turns out to have much less influence than in previous studies, and long-run inflation appears to have been a result mainly of exchange rate devaluations and of foreign price rises, whereas money was important only in the short run and the excess of wages over productivity or excess demand failed to have any impact.
The minor study is a survey of the determinants of inflation and hyperinflation in Latin America during the 1980s by Stefan de Vylder. 130 In a pedagogical fashion, the political background and the crisis of the state, as well as the external and internal determinants of the debt crisis are presented. Then the causal relations between political crisis, debt crisis, budget deficits and hyperinflation are traced and the quantity theory of money is applied to highlight the importance of the expansion of the money supply for hyperinflation.

Political Economy
'Political economy' may be defined as the study of the interaction of economic and political factors. What is to be included under this heading to a certain extent, is a matter of taste. Here, I have chosen to deal with works where political economy is the central component. This basically leaves a number of works on three Caribbean countries: Haiti, the Dominican Republic and Jamaica. l3l Mats Lundahl has examined the economic history of Haiti in political economy terms.132 An appropriate starting point is the early rulers: Toussaint r;Ouverture and Jean-Jacques Dessalines and their economic system which was largely shaped by political factors. 133 These rulers faced a double dilemma. On the one hand, Haiti had been recognized by no other nation. In particular, the threat of a French return was imminent. This required imports of war material. On the other hand, Haiti's rulers also had to ensure that their main backers, the new elite, obtained an appropriate income. Together, these two factors made them return to the plantation system instead of allowing the peasant agriculture which had begun to take shape during the wars ofliberation to spread spontaneously. Given the goals of export revenue and elite income, the transaction costs for the former system were simply lower than for the latter.
The plantation system received a blow from which it was never to recover with the land reform of 1809. One of the most peculiar features of Haitian political economy is the role of this land reform in the creation of the predatory state in the nineteenth century.134 The plantation system depended on forced labor, also after independence. Once the plantation workers had been freed, however, the elite could no longer extract an income from them since there was abundant land to squat on. This made the elite turn to politics. During the 1843-1915 period a pattern developed whereby politics turned into little else than a source of income for the elite and those aspiring to elite status. The state became a predatory one ruled by a kleptocracy. The presidency became a price to fight over for the various contending cliques and the country experienced more than a hundred more or less successful coups and uprisings during that period. Most on the twenty-two rulers from 1843 to 1915 concentrated simply on in-come and failed to spend enough on security. Hence, they were short-lived in office. Only one survived his entire term in office and retired peacefully.
The same pattern of clique competition for power for the sake of plunder also developed in the Dominican Republic where most presidents during the period between 1844 and 1916 tended to be as short-lived as in Haiti. This has been investigated by Mats Lundahl and Claudio Vedovato (Evald Nalin) who also examine how the same logic has continued to rule political life in both countries during the present century. 135 A particularly bad case of predation is to be found in the two Duvaliers in Haiti: Papa Doc and Baby Doc. The mode of operation of the former has been analyzed by Mats Lundahl who shows how Duvalier systematically destroyed the existing power structure and built his own, on elements that owed allegiance only to him personally and how this, in tum, both afforded him a greatly enhanced sense of security and unprecedented success in plundering the country through a number of innovative devices. 136 Altogether, the degeneration of politics during Haiti's entire history as a free nation has resulted in a hysteresis-like situation where this very history acts as an obstacle to change for the better both in political and economic terms. Predation and kleptocracy continue to dominate the scene. 137 The fact that Haiti has been a predatory state has created a number of popular responses that have served to escape the consequences of kleptocracy or to improve conditions. 138 Many of them were in the nature of protests. These mechanisms date all the way back to the colonial period. The plantation slaves simply escaped, for longer or shorter periods, when conditions became unbearable. The latter type of absence (petit marronage) was mainly of the 'voice' type, if we are to employ the terminology coined by Albert Hirschman, 139 while the former (grand marronage) was a permanent' exit' device. Grand marronage continued during the early independence period and exit remained the most common form of protest during the nineteenth century. At the same time, however, a perverted type of voice, by those bidding for power, developed: the endless coups. Voice as a genuine protest mechanism was not resurrected until the American occupation, which, however, also saw a great deal of exit. Voice continued to be the most important protest mechanism after the occupation, not least as practised by the trade union movement that was built up in the mid-1940s, only to be silenced again by successive repressive presidents, notably the two Duvaliers.
It is no exaggeration to contend that protests against the sitting governments have met with little success. As a result, the common man in Haiti has had to think of alternative devices through which to improve his lot. One such device has been informal and formal cooperation, ranging from family organization, patron-client relations, labor teams and rotating savings associations to caisses populaires and consumer and producer cooperatives, the majority of which have, however, been unsuccessful. Social security in Haiti is largely a matter of private, informal, initiative and ultimately, the best way to a better life leads out of the country, notably to the United States.
The political economy of the Dominican Republic has been examined in the doctoral dissertation of Claudio Vedovato (Evald Nalin).140 His point of departure is the notion of a predatory state and he starts by demonstrating that this notion fits the history of Dominican government well, reaching a climax during the regime of Rafael Trujillo between 1930 and 1961. Vedovato then turns to the structure of international trade in the Dominican Republic and argues that this cannot be understood unless attention is paid to the political goals of the state and the institutions it creates. The example he provides is that of sugar, where Trujillo had a strong personal interest and hence employed both the army and the state treasury to stimulate exports. The pro-plantation bias also had a negative side: an anti-peasant bias. The peasants have been more or less systematically discriminated against in the factor markets with the result that the foreign trade orientation chosen fostered inefficiency and lack of dynamism in the economy. The anti-peasant bias was also reinforced by the import substitution policy followed in the manufacturing sector. This was also the result of the personal economic interests of Trujillo and continued after his death as well. Seen from an international perspective, the record of the state in the Dominican Republic is rather dismal. The personal interests of its rulers have systematically distorted the economy and hampered development.
In another dissertation, where ingenious use is made of simple, but powerful, economic theory, Anders Danielson has examined the political economy aspects of the expansion of the public sector in Jamaica. 141 The main topic is how the choice of policy for raising public revenue has affected consumption, income distribution, investment and growth in Jamaica from 1962 to 1984. This is done with the aid of a three-sector (agriculture, manufacturing, services) neoclassical general equilibrium model. The book focuses on patterns of public finance during the period when Manley's People's National Party was in government: 1972-80. During the PNP period government expenditure led (and increased) and revenue followed during a period when incomes were falling. The distributional policy shifted incomes systematically in favor of wage incomes (having a lower savings propensity than capital incomes) which made national savings decline. This made it necessary to finance no less than 90 percent of gross investments from abroad in the late 1970s. Taxing the mining sector and redistributing the proceeds to groups with a high import propensity made imports of capital goods difficult, which in tum had a negative effect on growth. Finally, Manley's distributional policy, which patronized the PNP voters, also had the effect of increasing service employment at the expense of manufacturing and this produced deindustrialization.
Danielson has also employed an approach that builds on the notion of economic surplus, defined by him as the difference between GDP and the necessary costs of production (subsistence wages to productive, i.e. privately employed,142 workers plus depreciation of capital), to argue that taxation of mining has been used to finance the import substitution policy in manufacturing and that taxation of the latter sector, in turn, has been a means of increasing output and employment in (particularly government) services. 143 He has also shown that during the 1962-84 period, the surplus was to an increasingly used for consumption and not for investment. l44 Finally, he traces the stagnation of the Jamaican economy during the 1970s to the changed disposition ofthe surplus: in favor of middle-and high-income wage earners at the expense oflow-income wage earners and capitalists, which on the one hand means that the demand for imported consumption goods increased while that of capital goods decreased and on the other that the savings rate fell. 145

Concluding Remarks
What conclusions may be drawn from our survey of Swedish economic research on Latin America? In the first place, the coverage has been uneven, both in terms of countries and in terms of topics. While the northern Caribbean (Haiti, the Dominican Republic, Cuba and Jamaica), Mexico, parts of the Southern Cone (notably Chile) and Peru have attracted a lot of attention, there are no studies on Bolivia, Ecuador (other than in the regional context), Paraguay, Panama, Guyana, Surinam or French Guiana, and very little on CentralAmerica, Colombia and Venezuela. That some of these countries have been left out is not surprising. They are small and peripheral in many ways, not least economically. More surprising is the fact that Brazil has been studied comparatively little, but presumably this is due to the language factor: Portuguese is a language spoken by few Swedes.
The coverage in terms of topics leaves a lot to be desired as well. Studies of development strategies and economic policy dominate. Agriculture, manufacturing, trade and direct investment are other 'favored' areas. Some of the major lacunae are found in the field of distribution (excluding studies on agrarian structure), capital formation and stabilization policy other than inflation studies and works dealing with development strategies and economic policy in general. The reasons for this are not easy to find and attempting to discover any systematic causes may prove to be a somewhat futile exercise. Here, it is the personal interests of the researchers more than anything else, that is likely to determine the choice.
However, it should be noted that the vast majority of Swedish economists working on Latin America think of themselves primarily as economists and only secondarily as area specialists. As a result, they specialize by choosing a particular field in economics rather than by selecting a country of their predilection and this in tum determines the pattern oftopics as well. This also means that often the concentration of a particular study is more on the illustration of the workings of some economic principle than on some problem which is central to understanding how some Latin American economy works. This is most notable in the studies of foreign investment and technology transfer and in some of the works on manufacturing.
The quarter-century that we have surveyed in the present article -1969-93 -constitutes the period when economic research on Latin America was 'established'in Sweden. (Little of any importance was produced before that.) Hopefully, it will not be the GoldenAge, to be followed by gradual decline (and fall) but rather, when the 'definite' history of Swedish research on Latin America is written (if ever) it will go down in history as the twenty-five years that were crucial for opening up a fruitful new field for applied economic research in Sweden. It is of course in the nature of this exercise to be incomplete, partly because of the way in which I have chosen to draw the limits as to what should be included and partly because I am bound to have missed some contributions. For the former I assume full responsibility. For the latter I must ask the forbearance of those whose contributions have been unjustly left out. Needless to say, all the opinions expressed in the text are my own.

1.
Including business administration would add very little. The only major work during the period under review appears to be Linde's (1980) thesis on marketing, which in any case is not an area closely related to economics. (The study contains case studies from Colombia.) 2. This exception is Gerholm and Matthis (1970), co-authored by a social anthropologist and a physician. There are two reasons for haaving included this book: the approach is economic and historic, and the level of analysis is basically macroeconomic. In this context, three books by Sven Lindqvist (1969Lindqvist ( ), (1973Lindqvist ( ), (1974 should also be mentioned because of the influence they exerted on contemporary social scientists when they were published. Lindqvist is himself not a professional social scientist but an author of fiction who, has also produced a respectable body of partly analytical, partly documentary non-fiction works on the problems of the Third World. 3. Garcia, Cordero and Izquierdo (1987).

6.
Blomstram and Hettne (1981), (1984). The origin of the dependencia school-Latin American structuralism -is compared with the writings of the Swedish so-called Stockholm school in Buzaglo (1987).
13. See Lundahl and Vedovato (1991), for an explanation of how the differences in agrarian structure between Haiti and the Dominican Republic arose. Marner (1980) provides a historic perspective of agrarian structures in Latin America.

21.
The latter process is analyzed in Caballero Urzlia and Silva Perez (1975) and Albuquerque W. (1980). Chilean agriculture during the junta period is also dealt with in Izquierdo (1989c).

27.
Above all in Marner and Sims (1985), Spanish translation with an expanded chapter on the colonial period in Marner and Sims (1992). Marner (1978a) deals with the period from the mid-nineteenth century up to the present. Argentina and Chile are singled out for special treatment in Marner (1992a) which builds on the Marner-Sims book. See also Marner (1976a) and (1991a), for comments on existing literature on the subject, and Marner (1989a), for a discussion of immigration to Latin America in the context of the overall demographic development of Latin America.

28.
Migration inside and out of Haiti is surveyed in Lundahl (1979), Epilogue. Cf. also Lundahl (1980c). Lundahl (1983b) deals with migration to Cuba and Lundahl and Vargas (1983) analyze the emigration of Haiti to the Dominican Republic.

33.
Magnus Marner (1974a) has compared silver mining in Spanish America and the Swedish province of Lapland during the seventeenth century in terms of colonialism.

35
. Brundenius (1973a) picks up the theme of transfer pricing and profit repatriation in the context of the so-called Decision 24 of the Andean Pact, which ruled that foreign companies must gradually sell out shares to domestic investors over a specified period.
44. An article by Mauricio Rojas (1988b) deals with different approaches to the origins of industrialization in LatinAmerica, especially with the ideological elements of the debate. Rojas contrasts more or less stereotype models of' industrialization' with an approach that is based on comparative economic history.