Peasant Household Strategies Technologies and Diversification in Andean Fragile Lands

by Corinne Valdivia and Christian Jetté

Content

Peasant households in fragile environments

Risk and Diversification

San José Llanga, the face of diversity and dynamic technologies

Household strategies in San José Llanga: clusters and findings

The criteria

Strategies in 1993 and 1995

The Criteria

Table 1: Land use patterns in San José Llanga, 1993

Table 2: Cluster Variables in the Productive Group in San José Llanga: 1993 and 1995

Table 3: Income Sources Consumption and Diversity in San José Llanga: 1993 and 1995

Findings for Semi-arid Andean Production Systems and Recommendations

References
 

Recognition of the limits and increasing degradation of highly productive regions has drawn attention to the potential of fragile lands in semiarid and mountain regions of the developing world (Graham-Tomasi, 1991; Scherr and Hazell, 1993). A new approach to agricultural development and technology design is required for these regions (Scherr and Hazell, 1993). The semiarid and arid highland tropics of the Andean region are characterized by diverse agroecological and climatic conditions. Farmers have been able to persist in this setting for centuries through various forms of social and economic relations.

To contribute to the welfare of poor rural households the Small Ruminant Collaborative Research Support Program (SR-CRSP) initiated a research program in the central highlands of Bolivia in 1993. The purpose of the research was to develop a methodological framework for understanding the forces contributing to environmental degradation in crop livestock systems in semiarid, fragile environments (Ellis, J. 1993). An interdisciplinary research approach in the social and biological sciences was developed, with periodicity informing our assessment of producer strategies and technological innovations. Periodicity (Coppock, 1994) offers a dynamic image of the social, economic, and natural environments in fragile lands (Valdivia and Markowitz, 1995). Rather than focusing on the fixed attributes of users, the system of production is seen as intrinsically in flux, with windows of opportunity for certain technologies appearing and receding. The timing is the result of interactions between climate, institutions, economic and policy conditions, population trends and migration, as these interact with the decisions makers.

This paper focuses on peasant households strategies in an Andean community, and the social economic and environmental forces that shaped them in the last thirty years in their interaction with new technologies. The next section discusses the role of diversification in peasant household production systems in semi-arid environments, and in particular in the Andean region. A brief picture of the Andean peasant community where SR-CRSP research took place, and overview of three main technological experiences in the last thirty years are presented in the third section. Section IV identifies the strategies pursued by peasant households in this community using cluster analysis in two years of climatic perturbation, a year of slightly below average rainfall with frost (1993), and the third worst year of drought, delayed rains, and frosts of the last eighteen (1995). The last section discusses the findings and lessons for technology design and development to increase food and economic security in fragile semi-arid Andean crop-livestock systems.
 

Peasant households in fragile environments

Risk and Diversification
Low income peasant households in semi-arid and arid environments of the Tropics make their production and consumption decisions under uncertain and risky conditions, characterized by partial market integration and climatic perturbations (Fafchamps, 1992; Ellis, F. 1993). Losses in production and income have a direct impact on food security in low income rural households. Complex economic and social relations are developed to protect against risk, and to take advantage of an heterogenous resource base in agropastoral communities. Households engage in multiple economic activities that are not perfectly correlated to reduce risk ( Walker and Jodha, 1986; Reardon Delgado and Matlon, 1992; Bromley and Chavas, 1989). Risk is managed ex-ante by diversifying the economic portfolio to smooth income throughout the year, and ex-post through loss management strategies aimed at smoothing consumption. Asset liquidation, income transfers, off-farm employment and temporary migration are examples of the latter (Morduch, 1995; Townsend, 1995; Reardon et al., 1992). Livestock play an important role in managing risk and securing food and resources for investments (Reardon et al, 1992; Valdivia et al. 1996).

Although diversification may reflect the struggle to survive in a risky environment, it may also be an indicator of a dynamic agriculture (Reardon Delgado and Matlon, 1992, p.268) where resources are fully utilized in markedly seasonal environments (Valdivia et al, 1996; Ellis, 1993). Others find that new commercial activities (cash crops) lead to specialization as von Braun Hotchkiss and Immink (1989), found in the humid tropics of Guatemala.

The Andes
Peasant households in the Andean region have a long tradition of diversifying (Martinez-Castilla, 1992; Kervyn, 1988; Hopkins and Barrantes, 1987) in their adaptation to environmental perturbations, historical processes (Cotlear, 1989; Kervyn, 1988; Jetté, 1993), and diverse agro-ecological conditions (soils, topography and climate). Studies of technological change in peasant economies of the Peruvian mountains show that even with more intensive agriculture and higher incomes there is no tendency to greater specialization in more profitable crops and activities (Kervyn, 1988). Non market relations such as reciprocity, kin relations and networks (Alberti and Mayer, 1974; Carter and Albó, 1988; Caro, 1993) play an important role in household persistence by facilitating access to land, labor and animals without cash outlays (Markowitz and Jetté ). These relations also contribute to consumption smoothing through income transfers when idiosyncratic risk events take place, as found in other regions of the world (Rosenzweig and Binswanger, 1993).

San José Llanga, the face of diversity and dynamic technologies
San José Llanga is an agropastoral peasant community, 116 Km south of La Paz (Bolivia's capital city). Located in the Central Altiplano, between 3,725 and 3786 mts. above sea level, is an area of high degree of abiotic perturbations such as periodic drought, frequent frosts, occasional floods and seasonal wind erosion (Washington-Allen, 1993). Mean annual precipitation between 1943 and 1990 at Patacamaya Research Station was 402 with a coefficient of variation of 31%, and a mean annual temperature of 10.4 Celsius. The closest weekly market is 16 km away. Approximately one hundred households distributed between six human settlements live in San José: Callunimaya, Sabilani, Barrio, Tholatia, Incamaya and Espiritu Wilki. Land use distribution of the 7,200 Has in 1993 consisted of 48% covered by native grasses, 31% in fallow, 6% with cultivated forages, 5% with food crops, and the rest, 10%, occupied by buildings, rivers, roads, and uncultivated land (Massy and Valdivia, 1995).

The agropastoral production system combines sheep and cattle herding on natural rangeland with crop production. The main food and forage crops are potato, quinoa, barley and alfalfa. Several varieties indigenous and new, of these and other crops are planted in spatially dispersed small plots of land. Fallow agricultural land, crop residues and forage crops are important supplements to grazing (Valdivia et al., 1995). Potato, quinoa and llamas are the main indigenous crops and livestock in the region. Criollo (adapted) sheep and cattle, together with barley cultivation, exist since the colonial period, and progressively displaced llama production.

San José is 18 km away from one of Bolivia's agricultural experiment stations. During the sixties extension workers introduced new potato varieties, and popularized the use of tractors, fertilizers and pesticides. New breeds of sheep along with alfalfa were introduced to improve both meat and wool production. Criollo breeds are smaller and more resilient than the improved. As Bolivian textile firms went bankrupt during the eighties wool became less important.

Although cattle have been in the village for almost as long as sheep, their purpose in the past was primarily as draft animals. A severe drought in 1983 convinced many farmers of the need for irrigation and diversification into dairy and alfalfa production. Three events facilitate the adoption o dairy technologies in the late eighties: the introduction of an extension program that organizes production and milk delivery, Fomento Lechero (FL), the construction of a road from Patacamaya to SJL, and agreements with the parastatal dairy program, Planta de Industrializacion Lechera (PIL) to collect milk and support the price.

Precipitation varies between and within years, with very low rainfall from April through August. Frosts are considered the main danger for food crops. Potato cultivation has one possibility in two of completing its normal cycle, while quinoa has six possibilities in ten (Le Tacon et al, 1992). Notable differences in the incidence of frost risks have been found within the community. The construction of a 23 km. irrigation canal by community members at the beginning of the 1980s to deal with drought, brings water from the Desaguadero river (the main Central Altiplano stream) to a small portion of San José and four communities at the end of the dry season. This is used for alfalfa crops which supplement grazing on natural rangeland. The water is causing salinity. Alfalfa is also grown in fluviosols along the banks of a small seasonal river that crosses the north part of the territory.

Household strategies in San José Llanga: clusters and findings
Two production and income surveys were conducted in the community, one in 1993 an agricultural rainfall year of 388.5 mm, and in 1995 with 241mm. Both were years with frost during the growing season, and delayed rains in 1995. The first survey was applied to a randomly selected group of 45 households. The second was applied to the same families, though six had migrated, and were substituted by six new families.

The criteria
Groups of producers with similar strategies based on access and quality of resources, labor and age of household members, wealth, off farm employment, and animal species were developed through cluster analysis(1). To identify distinct sets of household strategies, the following variables were selected:

1. Life cycle, captured through the age of male head of household, and family labor available (Valdivia and Jette, 1996); 2. Quality of land included planted and utilized forage hectares of alfalfa and barley; 3. Differences in livestock technology included number of improved sheep belonging to the family, including crossbred animals, and number of criollo sheep belonging to the family capturing indigenous technology; 4. Number of improved cattle that belong to the family, both adult and young animals, an indicator of commercial production and technology adoption; 5. Number of criollo adult and young cattle that belong to the family, reflect indigenous technology; 6. Wages, a strategy of off-farm diversification indicative of market integration; 7. Consumption, the sum of in-kind and cash income from food crops (potatoes, quinoa, wheat, grain barley, faba beans), sheep consumption and sales, as well as sheep milk, cattle milk sales, wages and other income such as handcrafts and sales of thola (wood shrub) and manure, to determine differences between groups regarding welfare; and 8. Net income from cattle through the sales of live animals, reflecting households ability to capitalize and reinvest. Land property was not included because even though property differences existed, access was similar at differing income levels (Markowitz and Jetté).

Strategies in 1993 and 1995
Cluster analysis grouped the households in two large sets, both in 1993 and 1995: the "elderly" and the "productive", with age and access to labor being the defining variables. Table 1 presents the land use characteristics of these groups in 1993. The "productive", households in their middle years with access to family labor planted more plots of food crops such as potatoes, quinoa, barley, faba beans and canawa. All households planted potato, the basic food staple of the Andes, combining indigenous and new varieties. Their purpose is to satisfy primarily their needs. Main differences between the "improved" (households adopting new technologies) and the "criollo" (those that use indigenous technologies), were forage area planted to alfalfa and barley, irrigated lands, and fallow fields in 1993. Studies in the area also found that fluviosols and irrigated land under cultivation are more representative of distinct producer strategies. Similar findings were obtained in Peru (Hopkins and Barrantes, 1987). These lands are almost exclusively destined to forages in San José. This is consistent with the fact that animal numbers and species reflect more accurately the intensity of agricultural activity and wealth levels. The majority of the households in the "improved" group were located in Tholatia, while the "criollo" group was located in Espiritu Willqui, zone furthest from the main town and of poorer land resources and more unimproved range lands. The "elderly" for the most part have bequeathed their land and animals, but do plant potato (table 1). Standard deviations in parenthesis underline the variability surrounding land use patterns, especially range land, fallow crop land, and renting of land within the community.

Table 2 shows the means and standard deviations of the variables used in cluster analysis. The "productive" groups pursued different strategies in 1993 and 1995. Both improved and criollo sheep are the domain of women in the community in terms of decisions, labor and income control (Valdivia et al, 1995). Sheep are the basic source of protein (Murillo and Markowitz, 1995), and are sold to purchase food, school supplies and other household goods. Relative prices of improved to criollo were 1.38 and 1.7 in 1993 and 1995 respectively. An increase in improved sheep in both groups of "productive" farmers takes place between 1993 and 1995. The "improved" group of households increased the number of livestock, especially sheep, in absolute terms. The "criollo" group also increased the number of improved animals, though partial trade off with criollo animals took place. The most notable change in the "criollo" group is the increase in

Table 1: Land use patterns in San José Llanga, 1993
Variable/Groups The Productive Years The Elderly
Improved  Criollo Couples Singles
Potatoes (has) 1.1 (0.5) 0.7 (0.5) 0.6 (0.3) 0.6 (0.7)
Quinoa (has) 1.0 (0.8) 0.7 (0.9) 0.3 (0.2) 0.3 (0.4)
Food Crops (has) 3.3 (1.6) 2.5 (2.1) 1.5 (0.8) 1.1 (1.1)
Food Crops (plots) 6.2 (2.1) 5.2 (2.8) 4.4 (2.6) 2.8 (1.0)
Alfalfa (has) 3.3 (2.3) 0.8 (0.9) 0.6 (0.7) 0.3 (0.4)
Forage barley (has) 1.6 (1.2) 0.8 (0.8) 0.2 (0.2) 0.2 (0.2)
Irrigation (has) 1.4 (1.4) 0.3 (0.7) 0.4 (0.7) 0.9 (1.9)*
Land renting (has) 1.4 (2.6) 1.9 (2.3) 0.5 (0.8) 0.6 (0.7)
Rangeland (has) 5.6 (6.6) 6.1 (9.0) 1.0 (1.0) 3.1 (3.5)
Fallow (has) 11 (9.6) 5.3 (5.5) 2.0 (2.0) 6.7 (9.3)
N sample size 15 14 8 8
Zones: most frequent Tholatia (6) Espiritu Willqui (5) Sabilani (4) Sabilani (3)
*One household has five has and five have nothing. Standard deviations in parenthesis.
1. Clustering was undertaken using a SYSTAT cluster program, Euclidian with standardized variables. Distance metric is 1-Pearson correlation coefficient, complete linkage method (Farthest Neighbor). A rectangular data matrix was used to compute Normalized Euclidian Distances between objects. A dendrogram was constructed from which four clusters were identified with the 1993 data and five clusters with the 1995 survey.