CHAPTER 4. RURAL SPACE AND THE TERRITORIAL DIMENSION OF DEVELOPMENT IN THE MERCOSUR COUNTRIES - JOSE GRAZIANO DA SILVA

Introduction^[38]

Beginning in the 1960s, inspired by the ideas of the Economic Commission for Latin America and the Caribbean (ECLAC), there emerged a number of attempts to achieve integration among the countries of Latin America. The first of these was the Latin American Free Trade Association (LAFTA), signed in 1960, which sought to create a free trade area through the rapid elimination of trade barriers and tariffs between countries in the region. The problems encountered in the negotiating process, arising from differences between the signatory countries, led to the creation, in August 1980, of the Latin American Integration Association (ALADI), whose objective was to form a Latin American common market in a gradual, progressive manner. Within ALADI, MERCOSUR (the Southern Common Market) came into being as one of the regional subgroups.

Initial discussions on the creation of MERCOSUR took place in the first half of the 1980s between Brazil and Argentina. Several commitments were formalized between the two countries, culminating in the Treaty of Asuncion, which, on March 26, 1991, was also signed by Paraguay and Uruguay. In the second half of the 1990s, Bolivia and Chile began negotiations to bring about their entry into the bloc, an effort resulting in what has been called the expanded MERCOSUR - which, however, has still not been fully implemented.^[39]

The first stage of the integration process among the current members of MERCOSUR was the creation of a free trade zone, aimed at eliminating tariffs between the member countries by the end of the 1990s.^[40] The second stage was the establishment of a customs union process setting forth a common external tariff for non-member third countries. However, based on the problems encountered, lists of exceptions to the common external tariff were drawn up for products considered “sensitive”. These were to be effective until January 2006, at which time a common market among the original signatories to the Treaty of Asunción (Argentina, Brazil, Paraguay and Uruguay) is to be fully established. In order to achieve this, however, it will be necessary to first harmonize macroeconomic and sectoral policies, particularly fiscal and labour legislation (Montoya, 1998).

According to Zylberstayn and Jank (1998), MERCOSUR’s most important goal was to promote a more globalized and competitive regional economy by motivating member countries to reorganize their economies after years of protectionist policies. Some of the direct effects of the economic adjustments made in response to the Treaty of Asunción included an important impact on Brazilian wheat production, with some regional effects on corn, rice, beef, milk, onions, potatoes and wine. In Argentina, sugar, tobacco, timber, poultry and hog sectors were affected. Brazil’s advantage was in value-added products, while Argentina had an advantage in commodities.

In recent years, MERCOSUR was virtually paralysed by the severe crises that affected the largest economies in the bloc. The crises were the result of the rigid exchange policies adopted under stabilization plans undertaken by those countries. These included Argentina’s Cavallo Plan of 1991, which established parity between the peso and the dollar; and Brazil’s Real Plan of 1994, which established a system of exchange bands. The devaluation introduced by Brazil at the beginning of 1999 and by Argentina at the end of 2000 (which allowed their respective currencies to fluctuate against the dollar), while aimed at major long-term macroeconomic convergence, introduced new and unpredictable complications in the difficult process of paving the way for a common market. The devaluation of the Brazilian currency completely reversed the comparative advantages obtained by Argentina - not only for manufactured products, but also for some of the main agricultural commodities - during the time the fixed exchange parity between its currency and the United States dollar was in force. Exports destined for Brazil accounted for the bulk of Argentine products during the time the system of fixed or controlled floating exchange rates in the two countries was in effect. The devaluation of the Real, therefore, played a decisive role in deepening the Argentine foreign exchange crisis between 1999 and 2001, contributing to the intense social and political crisis in which the country currently finds itself. The Argentine crisis is presently infecting the economies of the other MERCOSUR countries, making expansion of trade within the region the most important issue for the bloc. Moreover, given that a possible solution to the crisis is not yet in sight, it is extremely difficult to make predictions on the future of MERCOSUR.

The present study attempts to offer a balanced view of the recent transformations that are occurring in the rural spaces of the MERCOSUR countries and indicate some possible routes to a form of development that assigns greater importance to the territorial dimension in order to overcome the traditional divisions between agricultural/non-agricultural and rural/urban concerns. This chapter consists of five sections. Following the introduction, the second section focuses on conceptualizing rural-urban space and the spatial perspective in rural development. It will also set the scene for the importance of institutions in fostering local participation, while at the same time dealing with the complexity and difficulties of local participation and decentralization. The third section presents a brief description of rural economic activities (agricultural and non-agricultural) in the MERCOSUR countries. The fourth section reviews experiences with decentralization and focuses on fiscal decentralization, particularly in terms of the transfer of federal funds. The final sections present some final considerations and areas for future research.

Defining the concepts of rural and urban

Unfortunately, there is still great confusion in the use of the terms “agricultural” and “rural” - terms that are often taken as synonymous in the Latin American literature.^[41] In truth, the term “agricultural” refers to a sector of economic activity concerned with the cultivation of plants, animal breeding and directly related activities, such as soil preparation, fence building, etc. Agricultural activities are supplemented by industrial and service-providing activities, such as the manufacture of fabrics, machines and equipment, trade and service-providing activities in general. The term “non-agricultural” activities, far from representing a sector, is merely a way of aggregating activities not related directly to agricultural production per se.

At the same time, what is called “rural” is really a spatial/geographic category, which in no sense defines economic activities, because in rural areas activities can be either agricultural or non-agricultural. Moreover, as has been exhaustively demonstrated by recent research,^[42] agricultural activities are decreasing in rural areas in terms of the number of persons involved and the income generated, while non-agricultural activities, particularly those linked to the provision of services, are increasing. For these reasons, families living in areas defined as rural are increasingly abandoning exclusively agricultural activities and turning more and more to a diverse range of activities.

According to ECLAC (2001), the concepts of urban and rural populations employed in different Latin American countries allow for a delineation of three different groups:

1. countries that define urban centre not only in terms of population density, but also in terms of availability of public utility services (e.g. paved streets, electricity, water, sewerage, medical and administrative services, etc.), as is the case in Chile, Costa Rica, Cuba, Panama and Uruguay;
   
2. 
   countries that set a minimum population of 2 000 inhabitants as the standard for a locality to be considered urban, such as occurs in Argentina, Bolivia, Guatemala, Mexico and Venezuela; and finally,
   
   
3. countries that stipulate that a city be the administrative seat of the municipality and allow within this definition populations of fewer than 2 000 inhabitants (or 250 households), with no other requirement in terms of available public services; in these cases, the municipalities also periodically revise the urban boundaries of localities. Most Latin American countries fit into this category, including Brazil, Colombia, Ecuador, El Salvador, Haiti, Honduras, Nicaragua, Paraguay, Peru and the Dominican Republic.

In all three groups, the rural population is defined by a process of exclusion - i.e. as that population that does not reside in urban areas. However, in none of the MERCOSUR countries does the definition of rural make any reference to the predominance of agricultural activities.

Table 1 presents data on the overall population - urban and rural - for the MERCOSUR countries, as well as for Chile and Bolivia. In an attempt to provide better equivalence in the respective definitions of urban and rural population in the different MERCOSUR countries, there is also a separate breakdown of the population for localities with fewer than 2 000 inhabitants (the criterion adopted by Argentina), based on data from ECLAC (2001). In aggregate terms the population of MERCOSUR is estimated at nearly 200 million for the 1990s, with 156 million (79 percent) classified as urban and 41 million (21 percent) as rural. This last figure increases slightly to 22 percent when people residing in localities of fewer than 2 000 inhabitants are included.

TABLE 1
Urban and rural population in the MERCOSUR countries, 1991-96

Source: ECLAC (2001).

Considering the MERCOSUR countries individually, Paraguay is far and away the most rural, with 52 percent of the population residing in localities of fewer than 2 000 inhabitants. At the other extreme is Argentina, the most urban of the MERCOSUR countries, with a rural population of barely 13 percent. With the exception of Uruguay, in no country in the region is there a meaningful increase in the rural population if one includes all localities with fewer than 2 000 inhabitants.

Any criterion used to distinguish between rural and urban will inevitably be arbitrary, and will appear more so to the extent that those spaces tend to form a continuum from metropolitan areas to those predominantly agricultural areas considered exclusively rural,^[43] without counting areas of biological reserves (or wildlife refuges, as they are known in the United States). Changes in this situation over the long term are at least as important as what criterion is used to distinguish between rural and urban. From one demographic census to the next, rural areas have steadily been incorporated within urban boundaries, as a result of the expansion of cities during the period between censuses. Comparison between two censuses always “urbanizes” a subset of persons who previously lived in rural areas that have been transformed into urban areas in the period between one census and another, thus inflating the estimates of rural exodus for the period.

Rural development and agricultural development

Until recently, it was assumed that rural and agricultural employment was in decline in Latin America. It was also postulated that the smaller the rural population, the more developed the region would be. The Organisation for Economic Cooperation and Development (OECD) countries, after decades of depopulation of their farmlands and the severe concentration of their populations in large cities, began in the mid-1980s to design specific policies to avoid the desertion of their rural areas. In the meantime, in Latin American countries, the rural exodus and abandonment of small and medium-sized cities were accepted as inexorable.

However, as noted by Anderson and Leiserson in their pioneering work (1980)^[44], non-farm activities were expanding very rapidly in farming regions of Africa, Asia and Latin America as a result of agricultural development itself, meriting special attention in the design of rural and even urban development strategies. According to data from 15 developing countries (Brazil among them), by 1970 non-farm income already constituted between 20 and 30 percent of rural employment, and the percentage could have been as high as 30 to 40 percent if villages and small towns are included in the definition of rural areas. The authors note that these percentages should be taken as minimum figures since official employment statistics reflect “primary occupation” during a particular period (generally only the week before the census interview) and often ignore secondary or informal rural non-farm employment of small farmers and of women.

Klein (1992) also called attention to the fact that in Latin America, rural farm employment (in terms of numbers of economically active persons) decreased 0.8 percent per year in the 1970s, while rural non-farm employment^[45] increased 3.4 percent per year, faster than the average growth of GDP. In 12 of the 18 countries for which census data were available, rural non-farm employment was increasing more rapidly than overall employment, as was the case in Brazil; in eight countries rural non-farm employment increased more rapidly than urban employment.

In data from 12 nationwide Latin American country surveys in the 1990s, the simple average percentage of household income deriving from rural non-farm employment for the countries overall is 46 percent, while the average for the rural populations (weighted by the rural population within each country) is approximately 40 percent (Reardon, Berdegué and Escobar, 2001). This large share contradicts the traditional view in Latin America that equates rural incomes with farm sector incomes. It is a share similar to the simple national averages of 45 percent found for Africa and 35 percent for Asia in similar syntheses of household survey data. In Latin America there is little inter-country variation in rural non-farm income shares, ranging from 35 to 50 percent, and little systematic relation of these shares with country GDP per capita.

However, Reardon, Berdegué and Escobar (2001) find that poor households and poor regions in general often lack access to better-paying non-farm employment that could bring them out of poverty. Instead, the rural non-farm activities on which they depend are the equivalent of subsistence farming in that they are low-productivity, low-wage, temporary jobs with little potential for growth. A major challenge to policymakers, the authors point out, is to promote non-farm employment and income with a real possibility of alleviating poverty.

The authors also show that there are a variety of growth engines for rural non-farm development, some of which are locally endogenous, as when agricultural development and commercialization generate surpluses that spur local non-farm development. While agricultural development might have been the fruit of investments by local staples farmers, it is itself often “implanted” by investment from non-local entrepreneurs. Other examples of growth engines include a rural non-farm activity spurred by outside investors or the employment of local families in non-farm jobs in nearby rural areas or in cities.

The findings of the studies suggest two main implications for policies and programmes. The first is the importance of providing incentives for households to seek rural non-farm jobs and increasing the households’ ability to respond to such opportunities. The second is that, recognising the importance of rural non-farm employment engines from outside the rural sector, policy should mobilise resources (human and institutional, in addition to capital) with the ability to develop new types of projects in such secondary and tertiary sectors as tourism, recreation and environmental services. The authors also emphasise the importance of promoting rural non-farm employment within the broader perspective of land-use development and the rural economy as a whole, rather than with a narrowly agricultural bias. Agricultural development itself requires growth in manufacturing and services. Relying exclusively on agricultural development means condemning these regions to chronic poverty, marginalization and stagnation (Reardon, Berdegué and Escobar, 2001).

Research for the MERCOSUR countries shows an unmistakable growth trend in non-agricultural rural employment. Most of these jobs, however, are low-skilled and poorly-paid, particularly those related to the provision of personal services (such as paid domestic work), which constitutes the majority of jobs in countries with high income concentration such as Brazil and Uruguay. While there are no recent data for Argentina, Paraguay or Uruguay to confirm growth in non-agricultural income for rural families such as is occurring in Brazil (Graziano da Silva and Del Grossi, 2001), there are strong indications of such a trend based on the sharp fall in prices of agricultural products exported by the region and the increase in government transfers related to pensions and retirement (see Giarracca, 2001; Neiman, 2001; Piñeiro, 2001; Galeano, 1997; Morley and Vos, 1999; Molinas Veja, 2000.)

The results presented above concerning the composition and growth of rural non-farm employment in the MERCOSUR countries do not differ greatly from the general patterns found in other Latin American countries. However, it is important to bear in mind the heterogeneity of activities lumped into the general rural non-farm employment “sector”. Moreover, the phenomena that are driving the growth of rural non-farm employment activities vary by country and depend on the links between rural non-farm employment activities and other sectors of economic activity, not least of which is the agricultural sector itself. Anderson and Leiserson (1980) note the increased demand for non-food goods and services generated by growth of agriculture and rising rural incomes as one type of driver of rural non-farm employment growth. They also emphasise the importance in some countries of external markets for handicrafts and for the output of large-scale agro-industries.

In the case of countries such as Argentina and Brazil, the demand generated by the urban sectors independently of local agriculture can be decisive in the growth of rural non-farm employment. Brazil has large metropolitan areas in practically all of its regions that profoundly influence the flow of products and persons, whether from city to countryside or vice versa. The agricultural activities in a given region can be reconfigured by urban residents living near rural areas as they seek leisure, tourism and environmental preservation. This gives rise to another dynamic of rural non-farm employment creation based on what we have termed “new agricultural activities” such as fee fishing, hunting lodges, raising animals and growing ornamental plants.^[46] Many of those activities, which were previously little developed and geographically quite dispersed, have become true productive chains involving agro-industrial operations, personal services and relatively complex, sophisticated systems of distribution, communications and packaging.

In the cases of Brazil and Argentina, all of the dynamics cited above are present. However, they are not the main drivers of rural non-farm employment in regions where the rural agricultural population is relatively small, where cities are large and where a large portion of the farm-sector EAP lives in urban areas, such as one finds in central-south Brazil (Graziano da Silva, 1999) and in the Pampean region of Argentina (Neiman, 2001). Moreover, in the regions where the process of agricultural modernization has been most intense, farming generates very little demand for labour, and the labour it requires is usually skilled. This labour is provided by firms, located in nearby cities, which provide agricultural services.

The research carried out by Project Rurbano shows that farm activities generate less income in Brazil, and that there has been a decline in the number of agricultural households, since they are unable to survive on their income from agriculture alone. Not even the number of pluriactive households, where the members combine farm and non-farm activities, has been increasing. Considering the fall in income from agricultural activities, rural Brazilian households are (and have been) steadily becoming more non-agricultural, with their survival depending on social service transfers (retirees and pensioners) and on non-farm employment (Graziano da Silva, 1999). The evidence gathered from studies in the other three MERCOSUR countries points in the same direction (Giarracca, 2001).

A recent IDB rural finance strategy document (2001) acknowledges that the rural non-farm sector is an increasingly important part of the rural economy and represents a growing share of total rural income and employment. Much of the document highlights the need to develop financial services other than short-term credit that will specifically enhance the productivity and expansion possibilities of non-farm service, processing and manufacturing enterprises. The document’s main conclusion is that rural financial markets do not function properly in Latin America and the Caribbean and that the underdevelopment of financial markets has a negative effect on productivity-enhancing investments, income expansion and sectoral growth.

Local development and participation

The idea that poverty is the most serious impediment to sustainable development in late-developing countries is far from novel. Yet it would be a mistake to imagine that this is the only, or even the greatest, problem. The lack of social organization -especially with regard to civil society - has proved to be a barrier just as great, or perhaps even greater, than the poverty of rural populations. This is especially true at a time when globalization has highlighted local spaces as arenas of social, economic and political participation for organized groups.

The local sustainable development approach suggests that most issues could be better resolved by removing power from the state in the federal arena and empowering the local level.^[47] At times, optimism prevents a clear examination of who within this local society is organized, which interests are effectively represented and how they are actually organized. The conclusions of such an examination would not be encouraging: in the non-developed countries it is still typically the old, oligarchic interests that are well represented at the local level. Social actors emerging in the rural areas of those countries are still in their infancy, with no organized institutional form. Non-governmental organizations (NGOs) often constitute a poor alternative for bringing together and representing interests. Many are little more than ad hoc arrangements aimed at winning governmental and foreign financial support and employing half a dozen “dedicated professionals” who could find no other form of survival. Lack of infrastructure (transportation, energy and water supply, sanitation, educational and health facilities, etc.) and high levels of urban poverty in small and medium-sized cities leave rural development a low-priority issue for local governments. Even the World Bank (1996) has suggested that local elites are more interested than national elites are in preserving inequitable social structures. Local elites tend to dominate local decision-making, thus serving their own interests at the expense of the interest of the poor. For these reasons, there is considerable scepticism as to whether decentralization will automatically lead to policies that are more sensitive to poverty.

Transformation and globalization

As is amply demonstrated in the following section, the rural areas of the MERCOSUR countries have undergone substantial transformations in the last two decades. In essence, they are influenced by urban activities that have moved them beyond merely agricultural production. As a consequence of this increasing urbanization of activities in the rural milieu, current official criteria designed to distinguish between urban and rural have been put in check, altering the traditional agricultural/sectoral focus of rural policies. Territorial policies are gaining importance and giving rise to new proposals for regional development (Tendler, 1997).

The debate on rural and agricultural restructuring in the context of globalization has not managed to escape the danger of “a reification of the local”, primarily in the less developed countries. Graziano da Silva (2002) emphasises four fundamental differences in the impacts of globalization on MERCOSUR countries vis-à-vis developed ones: (a) the emergence of new forms of governance; (b) issues related to social organization; (c) quality-of-life issues; and (d) the place that those countries (and their agricultural sectors) occupy in the new international division of labour. In the developed countries agriculture is a less important sector economically in terms of production, employment and income-generation. Moreover, the participation of developed countries in international agricultural markets came about as a result of surplus products and through importing of specific foods regulated by bilateral agreements, with prices and quantities usually established through negotiation.

In the case of the MERCOSUR countries, agriculture is still highly important economically, in terms of both creating jobs and producing food. It provides a significant proportion of the income and revenues that countries need to pay for their imports, which have increased significantly as a result of globalization. Therefore, access to international agricultural markets is becoming increasingly important, both for countries that export a significant portion of their products (such as the countries that established the Cairns group) and for countries that have always depended on the importation of agricultural commodities for domestic supply and/or domestic price control, as is the case with Brazil and Argentina. Furthermore, given the pressures resulting from the General Agreement on Tariffs and Trade (GATT), the MERCOSUR countries were compelled in the 1990s to open their markets to imported agricultural commodities in order to avoid retaliation from the developed countries, which did not wish to have their agricultural and non-agricultural exports affected.

Space and development

The roots of the differences among agricultural sectors in Latin American countries in general, and in the MERCOSUR countries in particular, go back to the period prior to the adjustment and structural reform policies implemented in the latter half of the 1980s, but were intensified with the advent of those reforms and with institutional changes specific to the sector. A central element in these changes was the withdrawal of the state and the attempt to replace governmental policies and institutions with market mechanisms, a process that affected practically all key aspects of the sector, such as allocation of credit, support services for producers, price policies and land distribution. According to David, Morales and Rodríuez (2001), the productive and social changes that were intensified by the reforms are related to processes that include technological change and its impact on earnings; increased competition from abroad; reduction in the amount of cultivated land; changes in the productive structure; a drop in employment and increased concentration of land and income - all of which began more than two decades ago. Below we provide a brief summary of the important trends in the agricultural sectors of the MERCOSUR countries.

Productive restructuring and labour productivity

Latin America’s agricultural sector overall grew at an annual rate of close to 2.6 percent during the 1990s. In the MERCOSUR countries, the rate was far above this average (see Table 2), with particularly notable growth in agriculture in Argentina, Chile and Uruguay, which exceeded 4 percent per year. Mexico, another important agricultural country in Latin America, grew only 1.3 percent per year in the same period, due to its accession to the North America Free Trade Agreement (NAFTA).

Looking at the individual performance of the countries in the most important commodity categories, Argentina, Bolivia, Brazil and Paraguay showed the greatest growth in oilseeds, due to expanded soy production. In fruits and vegetables, Mexico, Chile, Argentina and Brazil accounted for most of the expansion in Latin America. In tree farms, the greatest progress occurred in Chile, Argentina and Uruguay. Livestock also showed marked growth, with significant expansion in land used for this purpose in Brazil and Chile. The growth was primarily the result of the introduction of significant technological changes (David, Morales and Rodrigues, 2001).

TABLE 2
Selected Latin American countries: Growth in agriculture, forestry, hunting and fishing, 1970-1998

Source: ECLAC’s Agricultural Development Unit (in David, Morales and Rodrigues, 2001).

The downward trend in the international prices of some products, especially those traditionally part of the Latin American market such as wheat, coffee, sugar and cotton, had a negative impact on many exporting countries and was responsible for the poor performance of family farm producers. For that reason, the commodity categories showing the greatest growth in the region during the 1990s corresponded directly to more modern, capitalized producers, while crops in the hands of small farmers showed either stagnation or reductions in terms of amount of planted area. David, Morales and Rodrigues (2001) argue that this accounts for the sharp drop in the number of small farms during the last ten or twenty years, as found in recent studies of the agricultural censuses in Brazil, Chile and Uruguay and studies of the rural sectors in Argentina, Bolivia, Colombia and Mexico. This trend intensified during the 1990s.

Between 1990 and 1998, agricultural production per economically active person in the sector increased by approximately 20 percent in real terms in Latin America as a whole, growing from 2 002 to 2 393 US dollars (in 1990 US dollars). This is equivalent to an average annual growth of approximately 2.6 percent, well above the 0.5 percent annual rate achieved by the other sectors, thus demonstrating that productivity in agricultural work increased much more than in other activities. The greatest increases in this parameter occurred in Argentina, Brazil and Chile (David, Morales and Rodrigues, 2001).

During the 1990s the MERCOSUR countries tended to specialise in those products that had increasing global demand. In Paraguay, oilseeds, for which global demand has been steadily increasing, comprise 49 percent of agricultural exports and 44 percent of total exports. Exports of cotton have fallen given decreasing world-wide demand for vegetable fibres during the 1990s. In Uruguay, production suffered due to the drop in world-wide demand for two major exports, beef and grains. Bolivia and Brazil recovered during the 1990s due to their ability to specialize in the export of oilseeds. Chile has remained strong in fruits and vegetables, which are in great demand on the international market, although Chile is very dependent on foreign sources for grains, oilseeds and beef. Argentina showed the worst results among the expanded MERCOSUR countries during the period, due to its ambiguous position in the horticultural market: on the one hand, Argentina is a big exporter of temperate fruits, citrus and fruit juices; on the other hand, in recent times, there has been a marked increase in the imports of tropical fruits and certain vegetables (particularly prepared or frozen ones). Also contributing to Argentina’s poor performance is its dependence on imported poultry - one of the agricultural categories that showed the greatest increase in global demand during the 1990s - while its position as a net exporter of oilseed cakes and meal kept that indicator from falling into negative territory. An additional effect of the reforms was to reduce the prices of agricultural inputs, which produced a huge increase in levels of imports of such chemical inputs as pesticides, herbicides and fertilisers (David, Morales and Rodrigues, 2001).

TABLE 3
Selected countries in Latin America and the Caribbean: rural EAP by country, 1980-2000

Source: Estimates and projections of CELADE (in David, Morales and Rodríguez, 2001).

As a consequence of these trends, the rural economically active population (EAP) has increased far more slowly than the urban EAP. However, as can be seen in Table 3, the rural EAP figure varies widely between countries. In Brazil, for example, the rural EAP has nearly levelled out since the 1980s, while Argentina and Uruguay showed a reduction of rural EAP in absolute terms in the last decade. In the other expanded MERCOSUR countries (Bolivia, Chile and Paraguay) rural EAP has increased.

Nevertheless, there are marked differences between the stabilization of the rural EAP in the 1980-2000 period and the agricultural EAP of MERCOSUR countries, which dropped from 19.6 to 15.6 million persons between 1980 and 2000, representing a reduction of approximately 2 million people per decade engaged in agriculture (see Table 4). There are great differences between the MERCOSUR countries in this respect. In Brazil, the agricultural EAP dropped sharply; in Argentina and Uruguay, there also was a small drop, while in the remaining countries the EAP increased. This appears to be a result of changes in productive conditions and in patterns in the consumption of agricultural products, as well as growth in various services and activities in the rural sphere unrelated to agriculture, such as tourism and construction.

TABLE 4
MERCOSUR countries: Estimates of agricultural EAP, 1980-2000

Source: FAO, FAOSTAT Database, Rome, based on household and census surveys (in David, Morales and Rodrigues, 2001).

Approximately 29 percent of rural EAP in Brazil was involved in non-agricultural activities. The most important sectors of non-agricultural activities are consumer services, industry, commerce, social services and the construction industry. The number of these non-agricultural rural workers has been growing rapidly since the 1980s and they have become majority residents in some states, including Sao Paulo, Rio de Janeiro and Rio Grande do Norte (Graziano da Silva and Del Grossi, 2001).

Neiman and Bardomás (2001), analysing the changes in supply and demand for rural employment in agriculture in Argentina, argue that the process of modernizing agricultural production has affected the structure, location and distribution of the workforce employed in agriculture. Modernization of agriculture has also been responsible in part for the creation of non-agricultural jobs in the rural environment, though linked to broader changes in rural areas overall.

Based on the last two population censuses conducted in Argentina, the total workforce engaged in agriculture grew by approximately 13 percent between 1980 and 1991, from 1.21 million to 1.37 million people. For the same period, the EAP of the entire country grew by nearly 20 percent and the rural population dropped from 17 percent of the total population in 1980 to 12 percent in 1991. Despite that absolute increase in agricultural employment, its share of total EAP continues the historic downward trend, although in this period the drop is less marked: from 12 percent to 11 percent of total employment for the nation.

Neiman and Bardomás (2001) explain the change in EAP by the increase in the number of agricultural workers living in urban areas, who held 85 percent of the nearly 165 000 new jobs created between 1980 and 1991 thus increasing the urban agricultural workforce from 21 percent to 29 percent of the sector’s total employment. At the same time the size of the agricultural EAP residing in rural areas remained virtually unchanged. During the same period, the number of workers employed in non-agricultural activities who lived in rural areas remained virtually unchanged, both in absolute terms and as a share of total workers living in rural areas.

According to Piñeiro (2001), in Uruguay the agricultural sector’s EAP constitutes only 15 percent of the country’s total EAP. Within that number, the largest group of persons falls within the category of rural wage-earners, making the country an exception within MERCOSUR, where family workers represent the largest category.^[48] Based on the most recent available data, the rural population in Uruguay has declined significantly (dropping from 17 percent in 1975 to 13 percent in 1985) and is relocating within the rural environment. Many people are moving to the departments bordering Argentina and Brazil, motivated by differences in cost of living and by job opportunities. The departments in the centre of the country, with extensive agricultural and livestock production, are losing population. Four factors are instrumental in the strong rural-urban migration in certain departments: (i) expulsion of family producers and their workers; (ii) improvements in transportation; (iii) concentration of the productive base, given that the new categories of exports are produced by a mere few hundred producers; and (iv) technical changes that reduced the demand for permanent workers and increased demand for seasonal workers (Piñeiro, 2001).

The trend towards a reduction in the demand for labour in the agricultural sector in Latin America in general and in the MERCOSUR countries in particular, as summarised in Table 5, is directly linked to production and technological changes that have occurred in the region. The increasing gains in work productivity, for example, together with the relative stability of the agricultural frontier, resulted in a drop in employment. According to David, Morales and Rodrigues (2001), the expansion in livestock and forestry, which are not labour intensive, meant a reduction in job creation, while other expanding categories (e.g. fruit and vegetable growing and poultry farming) are increasingly using contract agriculture, which relies on greater capital and also reduces employment.

TABLE 5
Selected countries of Latin America: Relative changes in the importance of paid agricultural workers and uncompensated and independent workers, 1990-1997

Sources: David, Morales and Rodrigues (2001), Neiman (2001), Piñeiro (2001) and Galeano (1997).

Among the consequences of the structural changes that have occurred in the agricultural sector, including greater differentiation in productive conditions between small and large producers and the drop in number of jobs, some of the most adverse results have been the increase in poverty and inequality in the rural world. According to various sources compiled by David, Morales and Rodrigues (2001), approximately 66 percent of the poor living in rural settings - 47 million people - are small producers, 30 percent are rural settlers without land and the remaining 4 percent belong to indigenous and other groups. Of the small producers at least 40 percent are small-scale farmers, with little or no access to credit, technical assistance or agricultural support services and little ability to purchase land. The persistence of poverty in urban areas in spite of programmes designed specifically to combat it reinforces the suspicion that any number of similar programmes aimed at rural poverty - including those meant to boost productive development - will also be inadequate or misdirected, unless the specific problems of land, credit, capital and support services are addressed.

Structure of agricultural farms in the MERCOSUR countries

Neiman, Sánchez and Berger (2001) compiled data from the agricultural censuses of Argentina, Brazil, Paraguay and Uruguay, in all cases conducted when MERCOSUR was in its infancy (Argentina, 1988; Brazil, 1996; Paraguay, 1991 and Uruguay, 1994). Based on these censuses, the MERCOSUR countries have more than 6.5 million establishments involved in agricultural production, of which half occupy fewer than 10 hectares of land, while nearly 90 percent have fewer than 100 hectares of total declared land (Table 6). These figures are dominated by Brazil, as nearly 85 percent of all agricultural units in MERCOSUR are located in Brazil, with an even higher percentage of the small farms found there. However, if one looks at units of more than 10 000 hectares, over half are Argentine. Paraguay’s characteristics are similar to Brazil’s, while Uruguay’s agricultural structure is closer to Argentina’s.

TABLE 6
Distribution of farms by size and country

In Argentina, there are an additional 42 864 farms without defined boundaries.
Paraguay’s agricultural census lists 7 962 producers without land.

Source: Neiman, Sánchez and Berger (2001).

The agricultural sector in MERCOSUR employs more than 23 million people, including family workers and permanent wage-earners in agricultural establishments. Given the existence of approximately 6.5 million productive units, there are on average 3.5 persons per establishment, not counting temporary workers. Argentina and Uruguay show a broader pattern of employment, since their establishments provide permanent work to on average approximately 2.5 persons; by contrast, Brazilian and Paraguayan establishments employ only one more permanent worker per establishment.

The labour structure of MERCOSUR’s farms is dominated by family workers, who make up nearly 85 percent of the workforce employed on a permanent basis. The differences between countries are striking: from the high end in Paraguay, where 92 percent of campesino agricultural workers are family workers, to a low of approximately 50 percent for Uruguay. The average number of family workers per farm in Brazil and Paraguay is approximately 3, double the number in each of the other two countries. One constant in the agricultural systems of the four countries in regard to this indicator is that the greatest presence of family workers is found in farms with between 10 and 100 hectares (with similar differences for this stratum between countries and among the respective national averages). By country, the relative weight of permanent paid workers is greater in Argentina and Uruguay, covering at least one third of the total labour force. In the other two countries, the ratio between permanent paid workers and the total workforce is approximately 1 to 5. Likewise, Uruguay and Argentina show the highest average of paid workers per farm, with approximately one per unit, while in Brazil and Paraguay the average is one paid worker for every three farms.

According to Neiman, Sánchez and Berger (2001), the number of permanent paid workers rises as the size of parcels increases, contrary to what occurs with family work. This is particularly striking on large farms in Brazil and Uruguay, where farms with more than 10 000 hectares hire around 45 permanent paid workers per enterprise. In all of the countries, farms larger than 1 000 hectares hired approximately 3 paid workers per farm as a minimum. Farms in the smaller categories did not, on average, hire more than one person.

Family work in agriculture varies in relation to the various productive and socio-occupational contexts in the region. The number of family workers employed in Brazil and Paraguay correspond (as in the region as a whole) to those workers’ overall representation in the workforce. In Argentina and Brazil, however, family work is concentrated in the larger units (between 10 and 1 000 hectares). Permanent paid workers make up only 10 percent of the stable workforce in the region, although this estimate is likely to be low, given under-reporting owing to irregular hiring situations and difficulties in census taking. Argentina and Uruguay have the highest figures for the category, with as much as 40 percent of the total permanent workforce, while in Brazil and Paraguay it is the largest establishments that employ a significantly higher number of wage-earners per establishment (Neiman, Sánchez and Berger, 2001).

In Paraguay the stagnant and backward agricultural sector that predominated until the 1960s was followed by an agrarian structure marked by the predominance of agricultural enterprises focused on conservative modernization and based on authoritarian order. The crisis in the agricultural-export model that began in the 1990s continues to extend and expand today, taking new forms within a new context of democracy. Given that agriculture is the most important productive sector in Paraguay, that crisis extends to the rest of the economy. According to the 1991 census half of all rural families are involved primarily in agricultural activities and 75 percent had access to family parcels, almost all smaller than 20 hectares. Forty percent of the 307 000 farms with less than 5 hectares controlled less than 1 percent of the total land polled. In addition, two-thirds of small units had cotton as a cash crop and thus suffered the effects of the cotton crisis (Fogel, 2001).

Units of greater than 1 000 hectares, on the other hand, represented 1 percent of farms but 77 percent of total land surveyed. These latifundia dominate the power structure. Fogel argues that the issue of land permeates all dominant groups within Paraguayan society, from those that control financial capital to the influential officials who own land and whose interests are linked to land. Large landholders are also dominant political actors. In Paraguay, land concentration is at the root of current and potential agrarian conflicts.

Prior to the 1990s, agricultural production in Paraguay was dominated by cotton (primarily a campesino activity) and soy (produced by units with a minimum of 50 hectares), which together represented 70 percent of total exports for 1989 and 37 percent of agricultural production. In the 1990s, there was a sharp drop in traditional crops destined for the domestic market and for consumption by the campesino population, and a simultaneous increase in agricultural production, primarily soy. Rural poverty expanded and the number of people emigrating to urban centres increased (Fogel, 2001).

Galeano (1997) suggests that during the 1980s, differentiation among campesinos became more acute in Paraguay and began to have an effect on the structure of agricultural employment. The number of independent workers fell from 63 percent in 1982 to 58 percent in 1992, while wage-earners, mostly labourers, grew from 15 percent to 24 percent. (Actual employees constituted less than 1 percent.) There was a simultaneous reduction in the number of unpaid family workers. All of these phenomena resulted from the expansion of agricultural modernization, with its corresponding exclusionary effect, reinforced by the extremely limited access to new land for an increasingly rootless campesino sector.

This increase in paid workers in recent years has two striking characteristics, as pointed out by Galeano. There is an increase in the number of paid non-farm work in outlying areas, particularly for the most uprooted stratum of campesinos and in situations in which labour demand, particularly in specific phases of the productive cycle, is strong. For the most part, paid non-farm activities tend to have high turnover rates.

Fiscal decentralization

The purpose of this section is to present a general perspective on the experiences with decentralization in Latin America, particularly in the MERCOSUR countries, with emphasis on fiscal decentralization in rural development programmes. Recent pressures experienced by Argentine provinces and by Brazilian states and municipalities in cutting expenditures illustrate the dilemma faced by current decentralization programmes in the MERCOSUR countries. On the one hand, decentralization stimulates political participation, particularly among groups that are marginalized at the federal level. On the other hand, the additional demands of these groups cannot be met, owing to the limits imposed by fiscal austerity policies in force or required by agreements with the International Monetary Fund.

Decentralization: A political or an economic choice?

Finot (2001) assesses the policy and economic aspects of two decades of decentralization policies in Latin America. The policy aspect focuses on shifting the provision of certain public goods from the national sphere to the subnational level. The economic aspect seeks instead to alter productive processes to increase their economical competitiveness. Finot identifies the democratic participation of citizens in decision-making as the basic condition under which decentralization can contribute to economic efficiency. He points out that as important as the decisions citizens might make regarding expenditures are those regarding the contributions they will make to support such spending - one aspect as yet not included in Latin American processes of decentralization. Thus while decentralization has contributed to important advances in coverage of social services and even political participation, it has created pressures on fiscal balance and has not contributed to reducing economic concentration.

There is consensus in the literature on the strong relation between democratization of Latin American countries and the decentralization process implemented in the last two decades of the twentieth century. As a general rule, the military regimes of the 1960s and 1970s were strongly centralized. As civil liberties began to be restored, the demand for economic decentralization was mixed - and, one might even say, confused - with the issue of political participation. In this respect, fiscal decentralization gained impetus and became an economic as well as a political target, and served as a way for citizens to have greater control over spending.

Table 7 indicates the degree of decentralization of government spending in the MERCOSUR countries, compared to Mexico, which is a fairly decentralized federal republic by Latin American standards. Argentina, Brazil and Colombia form the trio of countries with the highest level of decentralization in Latin America. According to the IDB (1997), in the last 15 years the amount of public expenditures managed by local governments (municipal or provincial) in Latin America has grown from 8 percent to 15 percent on average. In the case of Brazil, the ratio of central to subnational government net revenues (after transfers) went from 70:30 in 1980 to 58:42 in early 2001 (Villela, 2001). In a number of countries, however - particularly smaller countries with smaller populations, such as Paraguay - resources continue to be highly centralized in the federal government.

TABLE 7
Countries and level of decentralization

Source: IDB, 1997.

There are also many differences among the MERCOSUR countries in the types of decentralization processes undergone, particularly as regards the transfer of resources and responsibilities to intermediate-level governments or municipalities. At the height of Brazil’s democratization process that returned political power to civilians, the 1988 constitution approved the transfer of resources and responsibilities directly to the municipalities. Argentina, for its part, chose to strengthen the intermediate levels of government in the provinces, representing in reality a regionalization of the country.

A review of the distribution of responsibility for executing public services for 24 functions in 18 countries, conducted by the IDB (1997), indicates that in the more decentralized MERCOSUR countries (Brazil and Argentina) many functions are shared among different levels of government. Such is the case with social services, including primary and secondary education and health care, as well as public housing. In Paraguay and Uruguay the subnational level of government accounts for a smaller portion of total public expenditure, with the central government taking responsibility for the majority of services. In these cases, municipalities concentrate on the traditional functions of maintaining streets and overseeing markets, slaughterhouses and cemeteries.

Villela (2001) points out that central governments tend to retain decision-making powers for allocating resources for services and not for executing services. In 12 of 18 Latin American countries the allocation of funds for housing and primary education is determined exclusively by the central governments, while in 14 countries the allocation of funds for welfare services and primary health care is overseen by the central governments.

Brazil is the only country in Latin America to have a subnational value added tax. This tax represents the major source of revenue for its states, with 25 percent of tax revenues going to the municipalities. Problems arise from individual states having primary responsibility for regulating this tax, among them high compliance costs and the tendency to stimulate “fiscal wars” (tax competition) between the states as they attempt to attract investors. These problems are highly significant for the system, but nonetheless this tax has been instrumental in making Brazil one of the most decentralized countries in the world. According to Villela, all other Latin-American countries have problems financing intermediate levels of government. The provinces of Argentina, the departments in Colombia and the Mexican states are all extremely dependent on central government transfers.

In a paper presented as groundwork for the IDB’s Subnational Development Strategy, Bird (2000) emphasises that subnational finance and intergovernmental fiscal relations are a focus of concern in Brazil and Argentina and are becoming increasingly important in other larger countries. Decentralization is also being considered by smaller countries such as Paraguay, Uruguay, Ecuador and Guatemala. Bird (2000) highlights the example of Argentina, where the national government has a preponderant role in determining the rates and bases of taxes, and in assessing and collecting them. While proceeds accrue to the provinces, such taxes should be considered as central government taxes that are allocated to the provinces through transfers. Such an interpretation is particularly plausible because there is little connection between the amount transferred and the amount collected locally.

Magrassi (2000) analysed subnational investment needs for the Latin American countries as well as the response within financial markets. He found that the need for local investment capital is growing throughout Latin American countries, and that several Latin American countries are attempting to increase subnational borrowing in order to expand and diversify the resource pool available to finance infrastructure investment. The public and private credit sources he cites include public banks, municipal development funds, commercial banks (often through public second-tier financial intermediaries) and capital markets. Marked contrasts between the constitutional and legal systems of Latin American countries, as well as the distinct characteristics of domestic credit markets, determine the differences between demand and supply of subnational borrowing in each country.

To visualize this variability, Magrassi (2000) presents an index of subnational borrowing autonomy level, developed for 18 Latin American countries for the above-cited IDB report dedicated to decentralization issues. The spectrum encompasses higher degrees of autonomy in the large federal states (Brazil and Argentina) and in some unitary states (such as Colombia and Ecuador). Magrassi argues that the greater subnational borrowing autonomy in these countries derives from the possibility of contracting debt without central government authorizations as well as the availability of future tax-sharing funds and the possibility of pledging them to secure the debt. Other countries limit the extent to which the subnational level can make debt-financing decisions (as in Bolivia) or do not allow subnational borrowing at all (as in Chile).

Magrassi (2000) identifies Mexico, Brazil and Argentina as having sufficiently developed financial markets to provide for subnational borrowing enhancement strategies. Each of these countries accounted for almost 30 percent of the total public and private debt issuance in Latin America in 1996. He also points to the more developed legal and regulatory infrastructure of domestic markets in Brazil, Chile, Argentina, Colombia and Uruguay, where locally-based independent rating agencies exist that can perform risk assessment on government and private debt issues.

The IDB’s subnational development strategy paper (2000) expands on the various and at times divergent motivations for promoting decentralization. These include decentralization as a mechanism to better allocate and use public resources (attaining a better match between the goods and services offered by government and the preferences of the population), promote engagement and ownership by the community of local development programmes, advance greater accountability and better governance at a local level, and take into account the needs and preferences of different regions of each country. Critics often note that local autonomy for decision-making does not by itself guarantee improvements in public services, and may in fact even worsen them if the local political elite control the decision-making process or if the distribution of resources does not balance the delivery capacity among subnational jurisdictions. These problems provide arguments to proponents of centralization, usually those concerned with macroeconomic stability and fiscal discipline and often located in finance ministries.

The World Bank (1995) notes other difficulties. Decentralization may fail if not accompanied by sufficient powers of taxation or central government transfers. While in the longer term, decentralization may increase the effectiveness of government spending, initial start-up costs to cover additional professional staff, buildings and equipment can be substantial. Further, while unconditional block grants improve the flexibility of local decision-making, they also facilitate the control of funds by local elites. Block grants also may take away incentives to local governments to raise their own resources.

The IDB (2000) takes note of the different types of subnational governments. Table 8 shows that the average population of 7 155 MERCOSUR municipalities is over 30 000, compared to an average population of 12 000 in European Union municipalities. So while most MERCOSUR municipalities are small, others are relatively large compared with the European countries. Thus on average a Latin American municipality must provide services to three times as many people as its counterpart in Europe.

TABLE 8
Population and municipalities in selected countries

Source: Basic data from IDB (2000).

The phenomenon of regional imbalances is discussed in the IDB document (2000) referred to above. The “agglomeration economies” that are generated by concentrating economic activity in cities favour productivity, but also put a strain on infrastructure and the environment, problems which require extensive investments to resolve. In rural areas, not only are there fewer people to work, but the concentration of productive activities (which are usually limited to a few extractive or natural resource-based activities) makes diversification difficult and renders local economies vulnerable to market fluctuations. Existing regional imbalances waste the local potential for economic expansion, create migratory and social tensions among regions, foster an uneven distribution of wealth and sustain the perpetuation of disparities among the regions. These regional imbalances are apparent in the gap between the resources and capacities of subnational governments in wealthy regions versus poor regions. The huge differences in size between subnational governments are reflected in their institutional capacity. While some large municipalities have institutional capacities greater than those of many small countries, with financial capacities and sophisticated financial management to match, small municipalities may have no permanent staff at all.

Rural development programmes

Parker (1995) emphasizes that decentralization is a multi-dimensional process that proceeds with successes and setbacks. Decentralization initiatives are therefore subject to a continuous process of modification, reflecting changes in social, political and economic conditions. After reviewing a wide array of experiences with decentralized rural development projects, Parker recognizes the impossibility of designing a single strategy for decentralization. Instead, he highlights the importance of the political, fiscal and institutional elements of the decentralization process for different rural development strategies.

Van Zyl et al. (1995) emphasise the fundamental importance of community participation and of delegating decision-making authority to lower levels of government and other institutions. They highlight three recent developments that distinguish current decentralization processes from previous attempts. The first is the establishment and/or extension of democratic institutions in many countries, including the replacement of military regimes by elected civilian governments and the fact that many local government officials/mayors and council members are no longer appointed but elected. The second is the belated recognition of the importance of providing financial resources to decentralized institutions in order to facilitate enacting their powers and carrying out their responsibilities, given that the lack of such resources doomed many earlier decentralization efforts. The third is the expansion of participation in decentralization to a wider variety of institutions, including privatization of some services and the inclusion of NGOs and community organizations for service delivery, as well as the improved targeting of vulnerable groups.

Brazil is one of the most decentralized federations in the world following the adoption of the 1988 Constitution, which clarified the respective roles of the different levels of government. Purely local functions have been assigned exclusively to the municipal level. These include elementary education, preventive health care and intracity transport. The responsibility for public services that are national in scope, such as defence and foreign affairs, remains exclusively a federal function. The remaining functions are the shared responsibility of the federal and state levels, with the federal government setting norms and the states being responsible for the delivery of services.

Van Zyl et al. (1995) point out the often substantial variance between de facto assignment and de jure assignment, along with the federal government’s persistent direct involvement in purely local functions. However, they note that in relation to responsibility for development programmes in particular, decentralization (if correctly managed) has the potential for removing financial and managerial problems associated with the over-centralization of project implementation, which has been identified as a major constraint in earlier interventions in the northeast. Decentralization enables local communities to play a more active role in project selection and implementation.

The World Bank Northeast Rural Development Projects in Brazil, for example, which were reformulated in 1993, focused on small, client-driven subproject investments implemented by the communities themselves, with decentralized project management at the state and local levels and more transparent decision- making and accountability for project performance. Their implementation has provided five important lessons:

1. Decentralization of fiscal and investment decision-making from federal to state and local governments tends to result in more efficient project administration.
   
   
2. Decentralization of resource allocation and investment decisions to rural municipalities and communities should be accompanied by a clearly-defined and well-disseminated system of checks and balances to discourage the misuse of funds.
   
   
3. Participation in the financing of subprojects generates a sense of ownership and a willingness to share responsibility for the future operation and maintenance of project investments.
   
   
4. Beneficiary participation in the selection, execution, supervision and financing of project investments ensures that investments respond to a true need, generate cost savings and increase accountability at the local level.
   
   
5. Sustainability of project investments has greater potential when the municipalities and communities contribute to the subproject financing in a cost-sharing arrangement and when there is increased beneficiary participation (World Bank, 1994).

Final considerations

The new approach to local sustainable development has the undeniable merit of moving beyond the archaic urban/rural and agricultural/non-agricultural dichotomies. As we know today, agricultural activities are profoundly transformed by non-agricultural activities. Rural space is not just a space defined by its particular relationship to the land - and, in broader terms, to nature and the environment - but is profoundly linked to contiguous urban space.

Nonetheless, the local development approach presupposes a minimum of social organization that allows different social actors to function as true protagonists in changing the spaces they inhabit. Thus, local sustainable development should also be understood as a political development, in the sense that it permits greater and better representation of diverse social actors. When we speak of these actors, we are not merely referring to agricultural producers, varied as that group may be. We must also include those who inhabit the rural milieu or those from urban areas who simply maintain an idyllic point of reference for a new relationship with nature.

Growing demands and concerns regarding the management and conservation of natural resources are other important components in the strengthening of rural space. Here, as well, the organization of social actors can provide the impulse for participation and implementation of local development plans oriented to their interests. There continue to be many restrictions on the forms of participation and representation, due both to low levels of mobilization and to difficulties in adequately representing all of the social sectors involved. This creates operational and organizational biases resulting from local institutional structures and the decision-making power of the most affluent groups.

In the case of MERCOSUR, while actions directed exclusively at agricultural development were successful in increasing modernization in some parts of the member countries, this was not accompanied by corresponding gains in rural development. One of the main reasons for this is that only the technological and economic dimensions of the rural development process were considered important, while social and political changes, such as the organization of rural landless workers and small farmers, were relegated to a secondary role.

Furthermore, with globalization, the disparities that exist in the MERCOSUR countries tend to be exacerbated, both in regional terms and, in the agricultural sector, with regard to the relationship between the family farm sector and agribusiness. It is apparent that globalization is accelerating the social exclusion of those already in a disadvantaged position, thus accentuating the imbalances that characterise agricultural modernization. While the new approach to local sustainable development makes it possible to overcome old urban/rural and agricultural/non-agricultural dichotomies, existing rural social organization may not be adequate to provide new social actors an opportunity to participate fully in decentralized mechanisms.

Attempts to move forward in fiscal decentralization to provide greater autonomy to regional governments and municipalities in the MERCOSUR countries butt up against increasing restrictions imposed by federal laws aimed at imposing more stringent fiscal measures, under the pretext of needing to maintain fiscal responsibility. These measures involve limiting spending on the basis of actual capacity for tax collection by local governments, plus transfers received (particularly in countries where political decentralization is more advanced, such as in Argentina and Brazil).

The concept that rural poverty is not the only impediment to sustained development in underdeveloped countries is far from novel. The lack of societal organization in underdeveloped countries - especially in the poorest rural areas - is becoming increasingly recognised as an important barrier, in an environment in which globalization is bringing renewed recognition of the value of local spaces as arenas for political, economic and social participation.

Finot (2001) argues that a strategic approach would deepen political decentralization in the provision of infrastructure and basic services and upon this, the foundation for local development would be built. This would entail operationally decentralizing a supply system that guarantees all inhabitants equal access to a “social basket” of public services. It would not necessarily have to interfere with leaving political decisions on production at a centralized level (or even recentralizing) for the purpose of favouring economic competition.

Finot proposes replacing current systems of transfers with a territorial system (based on autonomous decisions on local development) and a social system (the minimum basket of services). Disbursements should be proportional to the relative effort of each community. Here, not only the tax contribution would be taken into account, but also the contributions people make (through their social organizations) in terms of work, materials and money to the provision of public goods.

From evidence available in the wake of some 20 years of experience with fiscal decentralization in Latin America, Villela (2001) concludes that in spite of decentralization’s failure to reduce regional wealth inequality, especially in large countries like Argentina and Brazil, it has had a positive effect in terms of territorial and social equity. The funds available to the less developed regions have increased and the coverage of basic services like primary health care has improved. Villela cites improved allocation efficiency and greater service coverage (while acknowledging higher unitary costs), noting that results were better in the health sector, where central government has retained more responsibility for funding allocation, than in other sectors such as primary education, where funding was significantly decentralized.

Villela (2001) calls attention to macroeconomic imbalance and insecurity which can threaten to offset the benefits of decentralization, especially the tendency for expenditures to grow faster than local revenues. This can produce a gap that either has to be covered by increased central government transfers or that results in deficits and high subnational indebtedness. Fiscal responsibility legislation designed to control fiscal imbalance, promote transparency and re-establish subnational fiscal discipline has recently been introduced in several countries, although it is not clear whether it has been successful in establishing a sustainable intergovernmental pact of fiscal prudence and an environment for growth and democracy to prosper. The ongoing Argentine crisis, to cite one example, gives one pause in this regard.

How well has decentralization worked for rural development in the MERCOSUR countries? Although the initial evaluation by the World Bank (1995) of its experience with decentralized rural development projects recognises that more research is needed to gain a clear picture of what works and what does not, the report suggests that it is already possible to extract some major lessons from the experience to date. Primary is the importance of appropriate design of the three critical elements - political, fiscal and institutional - and the harmony between them. Appropriate design of how the three elements are combined cannot be determined by a set blueprint, but instead will have to be devised by each country, taking into account local traditions, history, politics and social factors.

Future research questions

A first area of research involves the issue of capacity building at local administrative levels as part of the process of decentralization. Little research or evaluation has been carried out regarding increasing local capacity in human resources, particularly in terms of negotiating, project and budget management and attracting private investment. Such research is particularly important given the importance of decentralization within the concept of territorial-based development in MERCOSUR countries.

A second and essential area for research is the question of social organization and participation in decentralization. Of particular interest is the independent or autonomous organization of emerging social actors in the rural arena where non-agricultural activities and “new” agricultural activities are increasingly important as ways of generating work and income. On this point, it is worth emphasising activities related to conserving natural resources and the possibility of sustainable exploration of resources for leisure activities and rural tourism.

A third aspect that merits further research is the new relationship between the family farm sector and agribusiness. How have the traditional relationships of the minifundio and latifundio been altered as a function of the changes occurring in Latin American agronomy during the last two decades?

Bibliography

Anderson, D. & Leiserson, M. 1978. Rural enterprise and non-farm employment. Washington, DC, World Bank Series Discussion Paper (January).

Anderson, D. & Leiserson, M. 1980. Rural non-farm employment in developing countries. Econ. Dev. and Cultural Change, 28(2): 227-248.

Bird, R. 2000. Intergovernmental fiscal relations in Latin America: policy design and policy outcomes. Washington, DC, Social Development Division, IDB.

Chaloult, I. & Hillcoat, G. 1997. MERCOSUR e comercio agropecuário. Buenos Aires, IDB Department for Integration and Regional Programmes. Institute for the Integration of Latin America and the Caribbean.

David, B., Morales, C. & Rodrigues, M. 2001. Modernidad y heterogeneidad: estilo de desarrollo agrícola y rural en América Latina y el Caribe. In M. Beatriz de A. David, ed., Desarrollo rural en América Latina y el Caribe, pp. 1-88. Bogotá, CEPAL/Alfaomega.

Del Grossi, M. & Graziano da Silva, J. 2002. Novo rural - Uma abordagem iustrada. Londrina, Brazil, Instituto Agronômico do Paraná.

ECLAC (Economic Commission for Latin America and the Caribbean). 2001. Urbanización y evolución de la populación urbana de America Latina, 1950-1990. Boletin demográfico número especial, May. Santiago, División de Población.

Finot, I. 2001. Descentralización en América Latina: teoría y práctica. Serie Gestión Pública, 12. Santiago, ILPES-CEPAL.

Fogel, R. 2001. La Estructura y la coyuntura en las luchas del movimiento campesino paraguayo. In N. Giarracca, ed. Desarrollo rural: ¿Una nueva ruralidad en América Latina? Buenos Aires, CLACSO - Grupo de Trabajo Rural.

Galeano, L. 1997. La pobreza en el Paraguay rural. Santiago de Chile, FAO (Human Rights Report).

Giarracca, N., ed. 2001. Desarrollo rural: ¿Una nueva ruralidad en América Latina? Buenos Aires, CLACSO - Grupo de Trabajo Rural.

Graziano da Silva, J. 1999. O novo rural brasileiro. Campinas, Brazil: Instituto de Economía-Unicamp.

Graziano da Silva, J. & Del Grossi, M. 2001. Rural non-farm employment and incomes in Brazil: patterns and evolution. World Dev., 29(3): 443-454.

Graziano da Silva, J. 2002. Local sustainable development, globalization and restructuring in underdeveloped countries. Int. Journal of Soc. of Agric. and Food, 10(1): 64-79.

IDB (Inter-American Development Bank). 1997. Informe progreso económico y social de América Latina. Santiago, IPES.

IDB. 2000. Making decentralization work in Latin America and the Caribbean: a background paper for the sub-national development strategy. Document number GN-2026. Washington, DC, Sustainable Development Department, Social Development Division.

IICA (Inter-American Institute for Cooperation on Agriculture). 2000. The new rurality: sustainable rural development in the context of a new reading of rural reality. San José, Conceptual Document Series.

Klein, E. 1992. El empleo rural no agrícola en América Latina. Document 364. Santiago, PREALC/ILO.

Magrassi, M. 2000. Subnational investment needs and financial market response. Washington, DC, IDB Sector Study, Regional Operations Department.

Molinas Veja, J. 2000. El mercado de tierras rurales en Paraguay. Serie Desarrollo Productivo 77. Santiago, CEPAL.

Montoya, M. 1998. A matriz insumo-produto internacional do Mercosul em 1990: as desigualdades regionais e o impacto intersetorial do comércio inter-regional. (Doctoral Thesis). Piracicaba - SP. ESALQ/USP.

Morley, S. & Vos, R. 1999. Poverty and dualistic growth in Paraguay. (draft)

Neiman, G. & Bardomás, S. 2001. Continuidad y cambio en la ocupación agropecuaria y rural de la Argentina. In G.Neiman, ed. Trabajo de campo. Producción, tecnología y empleo en el medio rural. Buenos Aires, Ediciones CICCUS.

Neiman, G., Alvarez Sánchez, A. & Berger, M. 2001. El trabajo agropecuario en el MERCOSUR: tendencias generales y diferencias nacionales. In CICCUS, ed., Trabajo de campo: Producción, tecnología y empleo en el medio rural, pp. 31-54. Buenos Aires, CICCUS.

Parker, A. 1995. Decentralization: the way forward for rural development? Policy Research Working Paper Series. Washington, DC, World Bank.

Piñeiro, D. 2001.Trabajadores rurales y flexibilización laboral. El caso de Uruguay. In Giarraca, N., ed. Desarrollo rural: ¿Una nueva ruralidad en América Latina? Buenos Aires, CLACSO - Grupo de Trabajo Rural.

Reardon.T., Berdegué, J. & Escobar, G. 2001. Rural non-farm employment and income in Latin America: Overview and policy implications. World Dev., 29(3): 395-409.

Tendler, J. 1997. Good government in the tropics. Baltimore, MD, Johns Hopkins University Press.

Van Zyl, J., Barbosa, T., Parker, A.N. & Sonn, L. 1995. Decentralized rural development and enhanced community participation: A case study from northeast Brazil. Policy Research Working Paper No. 1498. Washington, DC, World Bank Agriculture and Natural Resources Department Sector Policy and Water Resources Division.

Villela, L. 2001. Fiscal decentralization in Latin America. Washington, DC, IDB Integration and Regional Programs Department. (draft)

World Bank. 1994. Brazil Northeast Rural Poverty Alleviation Program - Bahia. Document Report No. 4390-BR. Washington, DC.

World Bank. 1995. Toward sustainable production systems and rural poverty reduction. Agriculture and Natural Resources Department Dissemination Notes 2. Washington, DC.

World Bank. 1996. The political economy of democratic decentralization. Agriculture and Natural Resources Department Dissemination Notes 9: 1-4. Washington, DC.

World Bank. 2001. Rural development strategy and action plan for the Latin America and the Caribbean region. Washington, DC.

Zylberstayn, D. & Jank, M.S. 1998. Agribusiness in MERCOSUR. Building new institutional apparatus. Agribusiness, 14 (4): 257-266.