One Drop at a Time: Diffusion of Modern Irrigation Technology in the Lurín Valley, Peru

Anna Barrett Schickele 14 May 2013 Advised by Martin Carnoy and Roz Naylor

2

Acknowledgements A big thank you to my advisors Martin Carnoy and Roz Naylor, and to Walter Falcon as well. I’d also like to thank Frank Fukuyama, director of the CDDRL Honors Program; Larry Diamond, director of CDDRL; and Eric Kramon, a most excellent TA. Mary Sprague, in the Public Policy Program, also gave exceptional advice. I appreciate the support I have received from everyone at El Centro Global, especially Juan Sánchez and Ana María Romero Lozada, during my Stanford in Government summer fellowship and two subsequent research trips. Thanks to Jennifer Wolf for helping me refine my interview questions and prepare for fieldwork and to the folks at Undergraduate Advising and Research who helped me secure grant funding. I should also recognize my parents and sister, who were not happy when I chose to spend my winter break in Peru. Finally, thank you to the other members of the CDDRL class of 2013 who have formed a wonderful academic community.

Table of Contents Chapter 1: Introduction 1.1 Hypothesis 1.2 Outline of Thesis Chapter 2: Review of Literature 2.1 Technology for Agricultural Development - Irrigation Technology 2.2 Economic Rationality of the Poor - Origins of the Hypothesis - Rationality in Action: Examples from Peru - Limits of Rationality and the Behavioral Approach - Institutions and Collective Choice 2.3 Conclusion Chapter 3: Formal Model 3.1 Factors of Interest - Sources of Income - Quantity Produced and Prices - Costs - Time Constraint - Perceived Risk - Full Model 3.2 Limitations of Model 3.3 Motivation for Interviews Chapter 4: Case Study 3.1 Background - Poverty and Inequality in Peru - Agriculture and Water Resources - The Lurín Valley - Lahuaytambo 3.2 Study Design - Collaboration with El Centro Global - Interviews with Farmers and Fieldworkers o Questions o Sample o Coding Scheme - Supplementary Data Sources - Limitations of Study

4 Chapter 5: Findings on Individual Choice 5.1 Opportunity Cost - Interviews with Farmers - Interviews with Fieldworkers 5.2 Perceived Risk - Interviews with Farmers - Interviews with Fieldworkers 5.3 Installation and Maintenance Costs - Interviews with Farmers - Interviews with Fieldworkers 5.4 Education and Training - Interviews with Farmers - Interviews with Fieldworkers 5.5 Market Integration - Interviews with Farmers - Interviews with Fieldworkers 5.6 Discussion Chapter 6: Institutional Context and Collective Choice 6.1 Importance of Institutions 6.2 Evolution of the Comunidades 6.3 Evidence - Interviews with Farmers - Interviews with Local Leaders - Interviews with Fieldworkers 6.4 Discussion Chapter 7: Conclusion 7.1 Policy Recommendations - Integrate Smallholders with Markets - Provide Financial Services - Introduce Appropriate Technologies - Collaborate with Existing Institutions 7.2 Generalizability - Outside Peru - Outside Agricultural Sector 7.4 Further Research and Final Thoughts Works Cited Appendix I: Interview Questions Appendix II: Photos

Chapter 1: Introduction “Es una oferta dos por uno.” It’s like buy one get one free. –A fieldworker on using drip irrigation to cultivate during the dry season During the dry season—April through December—there is little green in Peru’s Lurín Valley. Smallholder farmers traditionally irrigate by flooding their fields with water from a network of canals. As a result, the valley’s water supply, which accumulates in small reservoirs during the rainy season, often runs dry before the rains begin again. El Centro Global para el Desarrollo y la Democracia, a Lima-based NGO working in this rural, mountainous area, promotes modern irrigation systems as a way to enhance environmental sustainability, farm productivity, and rural incomes. Despite these advantages, El Centro Global fieldworkers have found few farmers interested in these irrigation techniques. In this thesis, I use “modern irrigation” to refer to two broad categories of irrigation: sprinklers, which deliver water to crops through an overhead spray; and drip irrigation, which delivers water directly to a plant’s root through tubes lying on the soil or buried just beneath (see Figure 1). Equipment within these categories ranges from simple, homemade sets to industrial systems with electronic sensors and computer controls. Even the most basic forms control the amount of water applied, conserving a resource that is scarce on the western slope of the Andes. Moreover, the technology allows Lurín Valley farmers to grow a second crop each year, which enters the market during a period of high prices. However, the majority of farmers continue to use traditional flood irrigation even

6 in places where municipalities, NGOs, and comunidades campesinas1 have collaborated to install modern systems. Figure 1: Sprinkler and Drip Irrigation Systems

Sprinklers (left) and drip irrigation (right). Both systems control the amount of water delivered to the crop more precisely than traditional flood irrigation. Sprinklers, which deliver a spray of water, can be used with fewer crops than drips, which deliver water directly to the root. Photo credits: Anna Schickele and USDA Economic Research Service.

I first visited the Lurín Valley during June and July 2011, when I received a Stanford in Government fellowship to intern with El Centro Global para el Desarrollo y la Democracia. I was struck by how my Peruvian coworkers struggled to engage farmers in projects that “made sense” to my coworkers and me in terms of their economic benefits. This senior honors thesis has grown out of my interest in how farmers decide to participate in or abstain from these NGO projects. I have continued to collaborate with El Centro Global in order to carry out this research project, which has included two subsequent trips to Peru, in August and December 2012.

                                                                                                                1

Many Peruvians in rural highland areas belong to comunidades campesinas, or indigenous peasant communities, which hold communal property and play a large role in determining access to water. Congress of the Peruvian Republic, Ley General de Comunidades Campesinas.

7 My thesis research focuses on the factors that inform a farmer’s decision to implement technological innovations. Although El Centro Global carries out a variety of activities in the Lurín Valley, I narrowed my focus to the organization’s efforts to promote drip and sprinkler systems. These efforts form an important component of the organization’s work to improve rural incomes and also appear to entail clear benefits for participants. Specifically, I sought to answer: What factors affect the decisions of farmers in the Lurín Valley to adopt modern irrigation techniques? Hypothesis Drawing from a significant body of literature that uses a “poor but rational” approach to understand the behavior of the poor,2 I hypothesize that farmers may have rational reasons to reject productivity-enhancing changes. In formulating my research question, I initially focused on the opportunity cost of farmers’ time and farmers’ perceptions of risks. Farmers in the Lurín Valley divide their time among agricultural work, off-farm employment, family duties, and community obligations, so participating in an NGO-irrigation project may have a prohibitive opportunity cost. Farmers growing just enough for their households’ survival may hesitate to adopt an unproven new technology, even when a visiting fieldworker explains the system’s potential for improving farm productivity. As my thesis evolved, I sought to include other rational factors and incorporate them into a formal model. My model (fully described in Chapter 3) proposes that farmers choose to adopt modern irrigation systems when the expected value of adopting is greater than the expected value of nonadoption. It includes yields and prices, the cost of a new irrigation system, the allocation of time between farm and off-farm work, and perceptions                                                                                                                 2

Duflo, “Poor but Rational?” 366.

8 of risk. I use this model to inform my interviews with farmers and fieldworkers from El Centro Global, conducted in December 2012. I focused on finding qualitative evidence that these rational factors—in particular, opportunity cost and perceptions of risk—have significant influence on farmers’ decisions. Outline of Thesis I have tested my rational choice hypothesis primarily through interviews with farmers and fieldworkers in the Lurín Valley, conducted in December 2012. While I found some evidence indicating that opportunity costs impeded the adoption of irrigation, I found less support for the hypothesis that farmers perceive irrigation systems as risky. My results emphasized the importance of other rational factors: the cost of installing and maintaining a sprinkler system; knowledge, training, and human capital; and integration with the market economy. Finally, my fieldwork shows that the institution of the comunidad campesina plays a central role in determining water access and encouraging the diffusion of irrigation technology. In the next chapter I review literature on the importance of technology for agricultural development and literature that applies the rational actor hypothesis to the poor. Then I use this literature to build a formal model of farmers’ decisions regarding the adoption of modern irrigation systems in Chapter 3. In Chapter 4 I describe my primary field location and my methodology for testing my hypothesis. I present my findings regarding my individual choice model in Chapter 5, and then I discuss circumstances where the adoption decision may be seen as a collective decision in Chapter 6. In the final chapter, I discuss the implications of these findings and conclude.  

 

9

Chapter 2: Review of Literature In order to be successful, an NGO- or state-led project in agricultural development must bring together productivity-enhancing methods with an understanding of how the target population will respond to them. In this chapter, I highlight two bodies of literature: literature on the importance of technology for agricultural development and literature that applies a framework of rational decision-making to the poor. This literature motivates the formal model of Lurín Valley farmers’ decisions regarding the adoption of drip or sprinkler irrigation that I develop in the next chapter. Technology for Agricultural Development In the second half of the 20th century, the technological changes of the Green Revolution dramatically reduced global levels of food insecurity.3 Improved varieties of maize, wheat, and rice, combined with fertilizer use and modern irrigation techniques, allowed for increases in cereal yields, especially in South Asia and Latin America.4 The gains from the Green Revolution were unevenly distributed, accruing mainly to farmers with more capital and land. Nevertheless, Conway explains that these advances are proof that investments in agricultural research and extension yield high rates of return.5 Investment in agricultural research and development may be essential to overcoming “poverty traps.”6 In subsistence conditions, individuals have no margin of income to invest in their future, so poverty itself limits their ability to increase their incomes.7 In these cases, state or NGO programs that help subsistence farmers make productivity-enhancing investments can allow individuals and their families to escape                                                                                                                 3

Conway, The Doubly Green Revolution, 44. Ibid, 62. 5 Ibid, 40. 6 Sachs, The End of Poverty, 56; Abhijit V. Banerjee, Poor Economics, 11. 7 Sachs, The End of Poverty, 56. 4

10 poverty.8 Sen discusses poverty in terms of “entitlements,” a term he uses to define the commodity bundles an individual can acquire through production or exchange,9 and he insists that “long-run policies have to be geared to enhancing, securing, and guaranteeing entitlements, rather than some simple formula like expanding food output.”10 On a large scale, Timmer writes that the right investments can raise productivity in both the agricultural and nonagricultural sectors, forming a crucial “pathway out of poverty” and initiating a “structural transformation” to a modern economy.11 Irrigation Technology Access to adequate water is a prerequisite for high agricultural productivity, so attention to irrigation technology is critical to agricultural development.12 Irrigation, or the “artificial application of water to crops,” increases yields, allows for cultivation in the off-season, and makes agriculture possible in arid areas.13 There are three main forms of distributing irrigation water: surface (canals), subsurface (pipes or tubes), and overhead (sprinklers).14 Surface irrigation is the oldest, simplest, and most common method.15 The network of canals traditionally used in the Peruvian Andes falls into this category, while new, water-saving techniques use subsurface or overhead distribution. Burney and Naylor, using evidence from Sub-Saharan Africa, create a conceptual model for the adoption of irrigation systems. They suggest that adopters regularly reevaluate their decision to use a new irrigation technique.16 Farmers will abandon new                                                                                                                 8

Abhijit V. Banerjee, Poor Economics, 11. Sen, “Food, Economics, and Entitlements,” 191. 10 Ibid, 200. 11 Timmer, A World Without Agriculture, 61. 12 Conway, The Doubly Green Revolution, 258. 13 Sivertsen and Lundberg, “Farming Practices and Environmental Problems in an Arid Landscape,” 147. 14 Ibid. 15 Ibid. 16 Burney and Naylor, “Smallholder Irrigation as a Poverty Alleviation Tool in Sub-Saharan Africa,” 112. 9

11 systems if they perceive it worsens their situation or results in only “marginal” efficiency gains.17 An irrigation project contributes significantly to poverty reduction when “the adopter realizes significant efficiencies and is able to reinvest the subsequent labor and cost savings, starting up the ladder of increasing investments and asset accumulation.”18 In line with the Burney and Naylor’s predictions, past irrigation projects have yielded mixed results. Effective projects often “leverage” existing local institutions and occur along with the switch to high-value market crops.19 When successful, the gains are significant. Burney and Naylor explain that a project to install solar-powered water pumps in Benin not only increased consumption, but also initiated small institutional changes: more children attended school, and women received formalized land rights and access to financial institutions.20 Dhawan analyzes the returns to drip irrigation systems in India and emphasizes that these systems have both “productive” (yield-increasing) and “protective” (environmental sustainability) benefits.21 However, irrigation projects can fail for many reasons. Agriculture is the largest human use of water,22 and irrigation often entails large environmental and social costs.23 Burney and Naylor emphasize that distribution technologies are ineffective unless paired with water access and a pattern of use.24 Modern irrigation contributed to the success of the Green Revolution as many improved crop varieties required a more consistent water supply, but the global demand for water has increased as a result. Conway explains that the development of high-yielding grains that can grow year-round “has placed a high                                                                                                                 17

Ibid. Ibid. 19 Ibid, 115, 117. 20 Ibid, 117. 21 Dhawan, “Drip Irrigation,” 3779. 22 Sivertsen and Lundberg, “Farming Practices and Environmental Problems in an Arid Landscape,” 147. 23 Conway, The Doubly Green Revolution, 251. 24 Burney and Naylor, “Smallholder Irrigation as a Poverty Alleviation Tool in Sub-Saharan Africa,” 112. 18

12 premium on the provision of irrigation water in the dry season, when the potential yields are greatest.”25 Furthermore, waterlogging and depletion of nutrients are serious problems.26 Salinization often occurs when irrigation systems lack adequate drainage.27 Irrigation projects are often corrupt, and the benefits fail to accrue to the small farmers the projects aim to help.28 Quiroga studies irrigation project in El Salvador and concludes that the project led to the “immiseration of the poor,” who could not successfully compete for benefits with farmers who have more capital and land.29 An experiment in Kenya distributed simple, low-cost drip irrigation kits to farmers, but finds a low take-up rate.30 Kulecho and Weatherhead conduct a series of interviews with farmers to evaluate this project and find that unreliable water supply, the inability to market crops, uncertain financial return from the system, and mechanical problems were common reasons why farmers failed to adopt or later abandoned the technology. Although irrigation is not a panacea, modern irrigation systems are often the best way to provide reliable and sustainable access to water.31 Good irrigation projects can mitigate the negative externalities associated with irrigation use, but they must take into account water access, water distribution, water use, and the institutional environment in which the project takes place.32 The benefits from a well-planned and well-implemented

                                                                                                                25

Conway, The Doubly Green Revolution, 250. Ibid, 253. 27 Ibid, 253; Sivertsen and Lundberg, “Farming Practices and Environmental Problems in an Arid Landscape,”143. 28 Conway, The Doubly Green Revolution, 254. 29 Quiroga, “Irrigation Planning to Transform Subsistence Agriculture,” 186. 30 Kulecho and Weatherhead, “Adoption and Experience of Low-Cost Drip Irrigation in Kenya,” 436. 31 Conway, The Doubly Green Revolution, 251, 253. 32 Burney and Naylor, “Smallholder Irrigation as a Poverty Alleviation Tool in Sub-Saharan Africa,” 121. 26

13 project extend beyond the individual farmer, forming a compelling argument for state or NGO intervention to promote modern, sustainable irrigation techniques. Economic Rationality of the Poor Although urban visitors to rural villages often dismiss traditional practices as “backward” and “unscientific,”33 a significant body of scholarly work uses neoclassical economic concepts to explain the preferences and choices of the poor. The term “poor but rational” summarizes this approach. These studies, according to economist Esther Duflo, “systematically explore the structural constraints that poverty imposes on the decisionmaking of an unboundedly rational individual.”34 Origins of the Hypothesis The “poor but rational” hypothesis is rooted in the work of anthropologist Sol Tax. More than half a century ago, Tax coined the term “penny capitalism” to describe the economy of the 800-person indigenous community of Panajachel, Guatemala.35 He writes that the village “may be characterized as a money economy organized in single households as both consumption and production units, with a strongly developed market economy which tends to be perfectly competitive.”36 The economy, although constrained by a lack of resources, is as productive as it can be. Nobel-prize winning economist Theodore Schultz builds on Tax’s work by applying a formal neoclassical economic framework to traditional agricultural systems. In Transforming Traditional Agriculture, first published in 1964, Schultz argues that traditional agricultural systems are in low-level equilibria that achieve efficient allocation                                                                                                                 33

Scott, Seeing Like a State, 241. Duflo, “Poor but Rational?” 375. 35 Tax, El Capitalismo Del Centavo. 36 Ibid, 51. 34

14 of the given factors of production.37 Increasing productivity means adopting new technology, or changing the factors of production.38 Traditional farmers fail to do so if the new agricultural factors are not profitable on a small scale, if searching for new factors is prohibitively expensive, or if farmers lack the knowledge and skills necessary to use these factors or use them effectively.39 As a result, Schultz argues for investment in farmers’ human capital.40 Rationality in Action: Examples from Peru Economists Adolfo Figueroa and Daniel Cotlear apply this rational choice framework to the economic decisions of Peruvian peasants. In La economía campesina en la sierra del Perú, Figueroa argues that the economies of Andean communities conform to Schultz’s hypothesis: low productivity is not the result of inefficiency but of the low quality and quantity of inputs available.41 He suggests that farmers seek to minimize risk, rather than maximize profits, writing that “the peasant family seems to demonstrate economic behavior motivated by securing a certain level of independence and self-sufficiency.”42 The pursuit of food security at the family or community level may lead farmers to forgo productivity-enhancing innovations. Figueroa writes that “it is very probable that the uncertainty associated with agriculture limits the diffusion of modern inputs” such as fertilizers and irrigation technologies.43 A classic example of the behaviors Andean farmers use to minimize risk is the custom of holding lands at several different altitudes. Figueroa writes that farmers seek to                                                                                                                 37

Schultz, Transforming Traditional Agriculture, 37. Ibid, 132. 39 Ibid, 174. 40 Ibid, 175. 41 Figueroa, La Economía Campesina En La Sierra Del Perú, 127. 42 Ibid, 37. 43 Ibid, 93. 38

15 have lands at several different pisos ecológicos (ecological niches),44 appropriate for growing different crops.45 Spalding also describes this pattern: “A community attempted to control as great a variety of different kinds of resource areas as possible, thus assuring the access of its members to the goods upon which it depended and protecting against crop failure and famine by avoiding undue reliance upon one bit of land that might fail to produce in a given year while another nearby piece, with slightly different conditions, would bear a crop.”46 To this day, farmers cultivate between three and 12 small plots. Cotlear’s book Desarrollo campesino en los Andes collects surveys and interviews on economic activity in three different areas of the Peruvian Andes. He examines how both technological and institutional changes can catalyze economic growth in the Andes. He writes that “an appropriate technology for [Andean conditions] must be (a) more profitable than traditional technology, and, at the same time, not increase risk very much; (b) sufficiently divisible that it can be used in small parcels; and (c) characterized by a bias towards the substitution of inputs that are scarce among peasant families.47 Peruvian farmers, according to Cotlear, make rational decisions to adopt technologies that are profitable and appropriate for the land and capital they possess. Limits of Rationality and the Behavioral Approach Human decisions are not entirely rational. Tversky and Kahneman, who founded the field of behavioral economics, famously argue against the rational actors hypothesis. They write, “The deviations of actual behavior are too widespread to be ignored, too                                                                                                                 44

In the Andes, altitude plays a large role in determining temperatures and viable crops. Spalding writes, “In this part of the continent, so near the equator, climate and vegetation are far more a function of altitude than of latitude.” Spalding, Huarochiri, 13. 45 Figueroa, La Economía Campesina En La Sierra Del Perú, 37. 46 Spalding, Huarochirí, 17. 47 Cotlear, Desarrollo Campesino En Los Andes 24.

16 systematic to be dismissed as random error, and too fundamental to be accommodated by relaxing the normative system.”48 Much of their work involves identifying the limitations of rational decisions: limited ability to process information, bounded willpower, adherence to social norms, and interests that are not purely selfish.49 Bertrand et al. argue that poverty can exacerbate the biases that steer humans away from purely rational decisions. They state, “The poor may exhibit the same basic weaknesses and biases as do people from other walks of life, except that in poverty with its narrow margins for error the same behaviors often manifest themselves in more pronounced ways and can lead to worse outcomes.”50 A response to these criticisms has been to develop economic models that incorporate these psychological insights. For example, Duflo, Kremer, and Robinson use a formal model with hyperbolic discounting51 to explain why farmers in Western Kenya do not use fertilizer even though it is profitable to do so.52 They argue that farmers defer the purchase of fertilizer to a later time period, overestimating their willingness to pay in the future.53 Their model predicts that a small, time-limited discount can overcome this procrastination problem, and they use a randomized controlled experiment to show empirical results consistent with this hypothesis.54

                                                                                                                48

Tversky and Kahneman, “Rational Choice and the Framing of Decisions,” 252 Duflo, “Poor but Rational?” 375. 50 Bertrand, Mullainathan, and Shafir, “A Behavioral-Economics View of Poverty,” 419. 51 Hyperbolic discounting is a time-inconsistent model of discounting the future. An individual values the present much more than a time period a short distance in the future, but the difference in valuation between two more distant time periods is relatively small. Duflo, Kremer, and Robinson, “Nudging Farmers to Use Fertilizer: Theory and Experimental Evidence from Kenya,” 6. 52 Ibid, 6. 53 Ibid, 15. 54 Ibid, 15, 35. 49

17 Institutions and Collective Choice Another departure from the framework of individual rational choice results from the fact that many economic decisions are made collectively, through formal and informal institutions. North famously defines institutions as “the humanly devised constraints that structure political, economic and social interaction”55 and writes that they are as important to economic growth as is technological progress.56 Political scientist Ostrom uses a rational choice framework to examine institutions and collective action. Although some academics suggested that rational individuals would never contribute to collective institutions, she writes, “A substantial gap exists between the theoretical prediction that self-interested individuals will have extreme difficulty in coordinating action and the reality that such cooperative behavior is widespread, although far from inevitable.”57 Much of her work discusses the conditions under which cooperation is the rational choice. Meinzen-Dick also studies collective choice, with specific attention to water resources management. She writes that water institutions are necessary because of the public goods nature and scale of water resources.58 In practice, state actors, collective action among users, market exchanges, or a combination of these arrangements coordinate individuals’ actions to their mutual benefit.59 Conclusion In this chapter, I have reviewed the literature that emphasizes the importance agricultural technology, including irrigation systems, can play in promoting agricultural                                                                                                                 55

North, “Institutions,” 97. North, “Institutional Change and Economic Growth,” 120. 57 Ostrom, “Collective Action and the Evolution of Social Norms,” 138. 58 Meinzen-Dick, “Beyond Panaceas in Water Institutions,” 15200. 59 Ibid. 56

18 development and the literature that applies the rational choice framework of neoclassical economics to the decisions of the poor. Rational choice is by no means the only approach that could be used to analyze the choices of farmers in the Lurín Valley, but I have chosen this approach because it clearly presents the relationships between many variables that may enter farmers’ decisions. In the chapters that follow, I develop a rational choice model, test it, and discuss the elements in farmers’ choices better explained with alternative models of decision making.

19

Chapter 3: Formal Model Drawing from the literature presented in the previous chapter, which discussed irrigation technology as a tool for poverty reduction and presented the rational choice approach, I build a formal model of Lurín Valley farmers’ decisions regarding the adoption of drip or sprinkler irrigation. In this model I assume that farmers are rational actors who produce for market exchange and seek to maximize their total income in a given year. They will choose to adopt drip irrigation if the expected value of adoption is greater than the expected value of nonadoption. Factors of Interest In my model, a farmer’s calculation of the expected value of adopting modern irrigation includes farm and non-farm sources of income, the quantity he produces, the prices he receives for the crop, the costs of production, how he allocates his time between various activities, and how risky he perceives the new technology to be. I first explain each of the variables and then present the full model. Sources of Income Farmers seek to maximize their total income, which is the sum of farm income and earnings from non-agricultural labor. I represent farm income as a function of yields and prices, and I use the variable w to represent the hourly wage the farmer can expect to earn from non-agricultural labor. Quantity Produced and Prices For simplicity, I represent farm income as a function of a single crop (or a composite crop) and its market price. Farmers who do not use modern irrigation can only grow one crop per year, in the rainy season, given by yRain, nonadopt. Farmers who adopt

20 modern irrigation can grow two crops per year: one in the rainy season (yRain, adopt) and another in the dry season (yDry). Adoption of a modern irrigation technique increases productivity during the rainy season (yRain, adopt > yRain, nonadopt) in addition to allowing for off-season production. Prices for agricultural commodities generally rise before the harvest and fall when the product comes into market. In my model, I make the simplifying assumption that there are two prices, pLow and pHigh. Farmers who enter the market at the end of the regular rainy season, when the crop is abundant, receive the lower price (pLow), and farmers who harvest during the off-season receive the higher price (pHigh). Thus, the yearly agricultural income of a full-time farmer who uses modern irrigation is: yDry *pHigh + yRain, adopt*pLow While the income of full-time farmer who does not adopt modern irrigation is: yRain, nonadopt*pLow Costs The installation of a new irrigation system entails a significant initial investment. I use the variable cAdopt to model costs, and I assume that there is no cost of nonadoption except lower income.60 In addition, I assume costs are fixed and do not vary with the quantity produced. Time Constraint Farmers have limited working hours to spend on each source of income, so they can only devote a fraction of working time t (0 ≤ t ≤ 1) to farm work and 1-t on nonfarm work. A farmer who spends all of his working time (t=1) in agriculture earns the yield                                                                                                                 60

This assumption is not entirely realistic because farm costs also include inputs such as seeds, fertilizer, and labor. I justify this assumption by arguing that the most significant difference in costs between the adopters and nonadopters is the cost of the irrigation system.

21 times-price expression presented above, but these earnings adjust with the fraction of time spend in non-agricultural activities. Adopting a new technology is time-intensive, at least initially, so the farmer spends more time on the farm if he chooses to adopt modern irrigation: tAdopt > tNonadopt. The value of the time spent on non-farm work is the hourly wage the farmer can earn in such work, w. Total non-farm income is given by (1-tNonadopt)*w and (1-tAdopt)*w. Perceived Risk The farmer may perceive adopting the new technology as risker than maintaining the status quo. If the new sprinkler system works, his average income will increase, but if not, he will be worse off than he was before the adoption.61 To simulate a farmer’s perception of this risk to his income, I multiply farm income by the coefficient λ. The risk coefficient depends on each farmer’s preferences or perceptions. It falls between 0 and 1 (0 ≤ λ ≤ 1), with lower values of λ indicating greater risk. To show that farmers perceive adoption as riskier than nonadoption, I assume λAdopt < λNonadopt. For simplicity, I assume that nonagricultural sources of income are risk-free (λ = 1). Full Model Taken together, my model yields expressions for the expected value of adoption and the expected value of nonadoption: EV(adoption) = λAdopt* tAdopt*( yDry*pHigh + yRain, adopt*pLow) – cAdopt + (1-tAdopt)*w EV(nonadoption) = λNonadopt* tNondopt* (yRain, nonadopt*pLow) + (1-tNonadopt)*w Farmers will rationally choose not to adopt modern irrigation technology when EV(adoption)>EV(nonadoption), or when:                                                                                                                 61

In developing my model and planning for fieldwork, I focused on the perceived risk of using modern irrigation technology rather than the perceived risk of integrating into the market economy. I now believe this was an oversight on my part, and I discuss the consequences at the end of Chapter 5.

22 λAdopt* tAdopt*( yDry*pHigh + yRain, adopt*pLow) – cAdopt + (1-tAdopt)*w > λNonadopt* tNondopt* (yRain, nonadopt*pLow) + (1-tNonadopt)*w The model shows that adoption is rational when adoption is not perceived as risky (high λAdopt), yields under adoption (yDry and yRain, adopt) are high, costs (cAdopt) are low, dry season prices (pHigh) are high, and wages (w) are low. Conversely, nonadoption is rational when adoption is perceived as risky, yields under adoption are not much different from yields under nonadoption, the time required to adopt is considerable, and nonagricultural wages are high. Limitations of Model My model excludes several key factors and uses a rational choice framework that may have inherent flaws. First, a more thorough model might assume that farmers maximize the net present value of income over their lifetime, rather than in a single season. 62 The model would use a discount rate to show how much farmers currently value income they will earn in the distant future in comparison with present income. I might assume that farmers with low discount rates (those who place a relatively high value on future income) and young farmers (who have many working years remaining) would be more likely to adopt.63 I also model costs in a blunt manner: I assume a fixed cost corresponding to adoption of the new technology, and I ignore costs under nonadoption. A more realistic model might separate costs from irrigation from the costs of other agricultural inputs, and                                                                                                                 62

The net present value of a stream of income Y is given by YNPV= Σ Yt/(1+r)t, where r is the discount rate. The sum runs from t=1 (the year of investment) to t=n (the last year of work). 63 One convention is to use the market interest rate as the discount rate. There are no financial services in the Lurín Valley, but banks in other rural areas of Peru charge very high interest rates (28-40%). Using this high discount rate would indicate that farmers are unlikely to install an irrigation system that increases income in future years.

23 when appropriate, vary these costs with area sown or quantity produced. It might incorporate the possibility of financing the technology with a loan, which would have to be repaid with interest over a certain period of time. This model might also distinguish between the up-front investment in an irrigation system, incurred in the year of adoption, and ongoing maintenance costs. In addition, I fail to explicitly incorporate the effects of education and training. Not knowing how to install and use a new technology is a barrier to adoption. Any training program requires time, so my model implicitly accounts for training. A more thorough model might demonstrate that farmers with more education, in general, or agricultural training, in particular, learn to use a new technology more easily. Furthermore, a farmer gradually learns how to use the irrigation system, so the gains from installing a new irrigation system might increase over time. Finally, as earlier authors have pointed out, the assumption of rational decisionmaking is not perfect. However, this approach remains a respected starting point for investigation. To the extent that rational calculations shape the behavior of any humans, they also shape the behavior of the poor. This framework also can also be helpful in revealing where individuals deviate from the model. Simplifying the farmer decision in this way can yield valuable insights as to how the included factors shape the outcome of technological change. Motivation for Interviews In this project, I do not seek to solve the model or estimate the parameters of interest. Instead, this model serves as a framework for my case study of irrigation technology in the Lurín Valley. In particular, the model motivates the investigation of

24 two specific factors, opportunity cost and perception of risk, by showing how these factors can create circumstances under which nonadoption is a rational choice. My model shows that opportunity costs will be high (making nonadoption rational) if adoption is time-intensive (high tAdopt) or if nonfarm income is considerable (high w*(1-tAdopt)). Perception of adoption as very risky (low λAdopt) likewise makes nonadoption rational. I seek to justify my inclusion of opportunity cost and perceived risk in the model through qualitative interviews with farmers and fieldworkers. Furthermore, my interviews test the overall appropriateness of this model, allowing me to either refine it by including new factors or identify reasons why the individual rational-choice framework is not appropriate.

25

Chapter 4: Case Study “Land is scarce and prized, and water is often scarcer”64 –Anthropologist Karen Spalding’s description of Huarochirí, the province where I conducted my fieldwork In this chapter I give background information on my field site, starting with an overview of agriculture in Peru. Then I introduce the Lurín Valley and the district of Lahuaytambo. I describe my methodology: interview questions, sample selection, and analysis of collected data. I conclude by discussing the limitations of my study. Background Poverty and Inequality in Peru Peru has experienced strong GDP growth in the past decade,65 but the country remains marked by the unequal distribution of wealth and economic opportunities. Peru’s Gini coefficient was 48.1 in 2010,66 and 30.8% of the population of 29.1 million67 lives below the national poverty line.68 This inequality takes a regional form: the urbanized coast accounts for 70% of GDP while low productivity and high poverty rates persist in the inland sierra (mountain) and selva (jungle) regions.69 James Thurlow and his colleagues at the International Food Policy Research Institute call this disparity “one of the country’s most pressing socioeconomic problems.”70

                                                                                                                64

Spalding, Huarochirí, 13. GDP growth was 9% 2007, 10% 2008, 1% 2009 (due to the global economic contraction), and rebounded to 9% in 2010. World Bank, World Data Bank: World Development Indicators and Global Development Finance. 66 Ibid. In comparison, the United States has a post-tax Gini coefficient of .38. OECD, OECD.StatExtracts. 67 Ibid. 68 Ibid. 69 Thurlow, Morley, and Nin Pratt, “Lagging Regions and Development Strategies: The Case of Peru,” 1, 3. 70 Ibid. 65

26 Given that almost one-quarter (23%) of the population continues to live in rural areas and works in agriculture, rural poverty and food insecurity are key concerns.71 Many poor Peruvians do not get enough food to meet their daily caloric needs, and 16% of the population is malnourished with an average deficit of 220 calories.72 Since most indigenous Peruvians (who form 45% of the total population) live in the poorer inland regions, these issues disproportionately affect indigenous groups.73 Thurlow et al. insist that “coastal growth undermines the inland economy” and that the country “cannot rely solely on rapid national growth to generate broad-based poverty reduction."74 Agriculture and Water Resources In 2010, 16.8% of Peru’s land area was used for agriculture.75 The country produced agricultural products worth $11.8 billion in 2010, amounting to 8.5% of total GDP.76 Peru exported agricultural products worth $3.2 billion, and the single most valuable export was coffee, worth $888 million.77 However, Peru’s regional inequalities extend to the agricultural sector. Most of the value produced in agriculture comes from large, commercial plantations on the coast, not from smallholder farms in inland areas.78 Approximately 28% of Peru’s permanent cropland is equipped for irrigation.79 Peru has abundant freshwater resources in the Amazon, but since most of its population lives along the desert coast, it is the most water-stressed country in South America.80 In the Andes, there are often water shortages during the dry season (April through                                                                                                                 71

World Bank, World Data Bank: World Development Indicators and Global Development Finance. Food and Agriculture Organization, Perfíl de País: Indicadores de Seguridad Alimentaria. 73 USAID, Property Rights and Resource Governance Country Profile: Peru, 1. 74 Thurlow, Morley, and Nin Pratt, “Lagging Regions and Development Strategies: The Case of Peru,” vi. 75 Food and Agriculture Organization, FAOStat. 76 World Bank, World Data Bank: World Development Indicators and Global Development Finance. 77 Food and Agriculture Organization, FAOStat. 78 Thurlow, Morley, and Nin Pratt, “Lagging Regions and Development Strategies: The Case of Peru,” 3. 79 USAID, Property Rights and Resource Governance Country Profile: Peru, 3. 80 Ibid, 10. 72

27 December) and floods during the rainy season.81 In many areas, agriculture competes with the mining industry for water access, and mining has contaminated the water supply.82 Unsurprisingly, sustainable management of water resources has become a national priority.83 The Lurín Valley My project site is the Lurín Valley, located three to four hours south of Lima (see Figure 2). The valley contains 10 distritos, an administrative division similar to a U.S. county. One of the distritos is called Lurín, but the name Lurín more commonly refers to the entire valley. Elevation in the area ranges from sea level to 3600 meters in the district of San Andrés de Tupicocha.84 The lower part of the valley (the districts of Lurín, Pachacamac, and Cieneguilla) is heavily urbanized and forms part of the greater Lima metropolitan area.85 The rural, mid-to-high altitude districts belong to the province of Huarochirí. In Lurín, as in other parts of the Andes, the comunidades campesinas (peasant communities) have primary responsibility for water resources management.86 The comunidades evolved from kinship networks that have existed since pre-Inca times, and their roles have been formalized under Peruvian law. The comunidades, which contain 100-500 members, are further divided into irrigation committees (comités de regantes) of 40-50 members.87 Farmers do not pay for water access, but must help maintain the community’s network of canals and reservoirs. In a study commissioned by El Centro                                                                                                                 81

Ibid. Toledo, personal interview. January 17, 2013. 83 Bleeker, personal interview. August 29, 2012. 84 Alfaro Moreno and Claverias Huerse, Diagnóstico Socioeconómico De La Cuenca Del Río Lurín, 5. 85 Ibid, 8. 86 Sánchez Barba, personal interview. August 29, 2012. 87 Felipe-Morales, Diagnóstico participativo sobre la gestión del agua en la cuenca del Río Lurín, 7. 82

28 Global, farmers complained both that inefficient irrigation practices are widespread and that the irrigation committees distribute water unfairly and ineffectively.88 Figure 2: Map of the Lurín Valley

The Lurín Valley, unlike other Andean river valleys, has no permanent icecap; its river depends on an underground spring and seasonal rains.89 Water demand, for both agricultural and non-agricultural uses, has risen in recent years.90 At the same time, farmers claim that climate change has made seasonal rains less reliable.91 Awareness of the need for effective stewardship of water resources has risen as a result.92 In recent years, the comunidades campesinas have worked to increase water supply by constructing more canals and dams, components of the traditional irrigation system. Some comunidades, notably the community of Tupicocha in the district of San Andrés de

                                                                                                                88

Ibid, 4. Felipe-Morales, Diagnóstico participativo sobre la gestión del agua en la cuenca del Río Lurín, 2. 90 Ibid, 2. 91 Bleeker, personal interview. August 29, 2012. 92 Felipe-Morales, Diagnóstico participativo sobre la gestión del agua en la cuenca del Río Lurín, 2. 89

29 Tupicocha, have successfully introduced drip irrigation, but others demonstrate less interest in adopting these water-saving techniques.93 El Centro Global para el Desarrollo y la Democracia, a Lima-based organization, is currently conducting a “territorial development” (desarrollo territorial) project in the Lurín Valley in collaboration with another NGO, El Centro para la Investigacion, Educación y Desarrollo (CIED). Since 2009, fieldworkers from El Centro Global have engaged valley residents in a variety of projects to strengthen agricultural productivity, attract tourists from Lima, and improve local governance.94 The organization has helped farmers purchase improved seed varieties, supported the creation of a network of mayors in the valley (called a mancomunidad), and helped finance the construction of a fruit-processing plant. El Centro Global has also sought to improve water resources management throughout the Lurín Valley. However, it lacks the budget to finance the installation of new systems itself, and it has made a deliberate choice to provide training rather than “handouts.”95 The organization’s founder, former Peruvian President Alejandro Toledo, says that the organization seeks to “empower instead of give gifts.”96 Fieldworkers often discuss irrigation techniques with individual farmers, make recommendations, and help farmers learn to use new systems. El Centro Global is currently collaborating with Aquafondo, an organization that finances hydrological projects in the three river basins

                                                                                                                93

Sánchez Barba, personal interview. December 20, 2012. “CGDD firma acuerdo con el BID para ejecutar un proyecto en el valle de Lurin,” El Centro Global website. 95 Sánchez Barba, personal interview. December 20, 2012. 96 Toledo, personal interview. January 17, 2013. 94

30 surrounding Lima, to build demonstration plots where fieldworkers train farmers in the use of sprinkler and drip irrigation systems.97 Lahuaytambo My primary field site was Lahuaytambo, in the upper part of the Lurín Valley. Travel from Lima takes approximately three hours in private car and upwards of five hours by bus. About half of the journey is on single-lane, unpaved roads. The elevation ranges between 2000 and 4000 meters above sea level, with the largest town located at 3338 meters.98 The 2007 population was 837 people,99 who belong to three comunidades campesinas: Llambilla, Pariapongo, and Santa Ana. Lahuaytambo is one of the valley’s poorest districts, with 75% of the population living in poverty.100 The average annual family income was 7973 New Peruvian Soles (3065 USD101) in 2012, and families spend 79% of income on food.102 Per-capita food consumption is low and infant malnutrition is high (48%).103 Agriculture is the predominant economic activity and most income (65%) comes from agricultural work.104 At lower elevations, the principal crops are apple and quince (which are generally sold to the market), and at higher elevations, farmers produce alfalfa, potatoes and broad beans (which are more often consumed or traded with neighbors).105 Many families also have a small number of cattle or sheep.106                                                                                                                 97

Aquafondo, Gestión Eficiente del Riego Tecnificado Para Cultivos Rentables. El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 2; Alfaro Moreno and Claverias Huerse, Diagnóstico Socioeconómico De La Cuenca Del Río Lurín, 7. 99 Ibid, 7. 100 El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 9. 101 One New Peruvian Sol = .3844 USD as of May 11, 2013. 102 El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 7. 103 Ibid, 4. 104 Ibid, 6. 105 Ibid, 2. 106 Ibid, 2. 98

31 Most farmers in Lahuaytambo irrigate with rainwater stored in small reservoirs, which limits them to one growing season per year.107 The local municipality and the comunidades campesinas are working to implement irrigation projects, mainly dams and canals. Between 2002 and 2004 the communities of Llambilla and Pariapongo installed drip and sprinkler irrigation on some communal fields. A program in Peru’s Ministry of Development and Social Inclusion, FONCODES (Fondo de Cooperación para el Desarrollo Social) financed the installation before a change in law banned using public funds for private uses.108 The modern irrigation systems were not used and are now in a state of disrepair.109 Study Design I use a qualitative methodology to answer my research question regarding the factors affecting the adoption of drip irrigation in the Lahuaytambo. My primary data comes from interviews with Peruvian farmers and fieldworkers. I supplement these data with documents generated by El Centro Global, expert interviews, and relevant literature. Collaboration with El Centro Global As mentioned above, El Centro Global para el Desarrollo y la Democracia is a small, Lima-based NGO conducting a multi-pronged development project in the Lurín Valley. Former Peruvian President Alejandro Toledo founded the organization when he left office in 2006. Toledo is a Stanford alumnus, and through this Stanford connection, I received a fellowship to intern with El Centro Global in the summer of 2011. I have continued collaborating with the NGO to write this thesis. My coworkers at El Centro                                                                                                                 107

Ibid, 3. Sánchez Barba, personal interview. December 20, 2012. 109 To illustrate, the municipality spent 160,000 soles (Roughly 61,500 USD) on drip irrigation systems, irrigation canals, and a small dam in 2010. Ibid, 10. 108

32 Global supported my applications for grant funding through Stanford’s Undergraduate Advising and Research. They provided me with a place to work during two research trips to Lima and provided transport to the Lurín Valley on multiple occasions. My colleagues also accompanied me during parts of my fieldwork. Interviews with Farmers and Fieldworkers I spoke with 23 farmers and four fieldworkers, resulting in approximately five hours of data on a digital audio recorder. Questions I developed two sets of open-ended interview questions, one for farmers and one for fieldworkers. I began my interviews with a brief statement of my research, and then I moved from simple to more complex questions.110 Both sets of interviews were semistructured, meaning I asked all respondents to speak to a pre-defined list of topics, but I allowed for the conversation to flow naturally.111 This approach also allowed farmers and fieldworkers to raise topics or concerns that I had not previously considered. I began my interviews with farmers by asking basic information about their plots, the crops they grow, and the irrigation methods they use. I then asked them to describe their interactions with fieldworkers from El Centro Global and their participation in any of the NGO’s projects. Each of these interviews took about 10 minutes. My interviews with fieldworkers included questions about the history of El Centro Global’s efforts to promote drip irrigation and the workers’ personal involvement. I asked them to share their perceptions of farmers’ decision-making processes, and I

                                                                                                                110 111

Judd, “Questionnaires and Interviews: Asking Questions Effectively,” 245. Ibid, 261.

33 solicited reflections on El Centro Global’s work. These interviews lasted 20-30 minutes. See Appendix I for a complete list of interview questions. Sample I interviewed 23 farmers, who I located through “snowballing.”112 My colleagues at El Centro Global introduced me to the farmers they work most closely with, and through these interviewees, I was able to interview family members and neighbors. A drawback of this method is that it favors respondents with social connections and those who have a positive view of El Centro Global.113 To mitigate this bias, I asked my coworkers to introduce me to farmers who had less positive attitudes towards modern irrigation systems and who had previously declined to participate in the organization’s projects. Furthermore, I recruited some subjects by approaching them in Lahuaytambo’s streets without introduction from El Centro Global. Table 1: Descriptive Statistics for Farmers Basic Data Sample size 23 Modern irrigation (drip, sprinkler, or combination) Women Mean interview length Age Oldest age Youngest age Median age Comunidad Campesina Llambilla Pariapongo Santa Ana Non member Household Structure Mean household size

4 9 10 min 67 23 50 10 6 1 5

Farm Characteristics Most common crops Potato 17 Wheat 14 Broad beans 13 Peas 12 Median number of crops 5 No connection to market 13 Education* Finished secondary school 10 More than secondary school 4 *Only 18 interviewees were asked this question Project Participation Knowledge of NGO 16 Participation with NGO 6

3.13  

                                                                                                                112 113

Weiss, Learning from Strangers, 25. Ibid, 29.

 

34 Some descriptive statistics for my sample of farmers are given in Table 1. Although my sample is not representative of Lahuaytambo’s population, my interviewees possess many of the characteristics typical of farmers in the area. The median age of respondents was relatively old (50), reflecting an area where many youth have migrated to Lima.114 Few respondents had children living at home. The most commonly mentioned crops are also those El Centro Global has identified as main crops in the region: potatoes, wheat, broad beans, and peas.115 Women are underrepresented in my study, forming only nine out of 23 subjects. My sample is also better educated than average; about 56% of the sample finished secondary school, while the figure is only 49% for the general population.116 I have ten respondents from the comunidad campesina of Llambilla, six from Pariapongo, one from Santa Ana, and five who are not members of a comunidad. Unfortunately, no data is available for comparison with the general population. I also conducted interviews with four employees of El Centro Global who are involved in varying components of the Lurín Valley program. This sample is not large enough to present descriptive statistics. Coding Scheme Following my fieldwork, I transcribed, translated, and coded my interview data. I coded in several passes, moving from descriptive to more interpretive codes.117 I began by tagging each speaker as a farmer or fieldworker. I then marked all the individuals who held some official leadership position within the community. Then I identified each passage where the speaker discussed a form of irrigation and applied uncontroversial,                                                                                                                 114

Alfaro Moreno and Claverias Huerse, Diagnóstico Socioeconómico de La Cuenca Del Río Lurín, 25. El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 2. 116 Ibid, 9. 117 Miles, Qualitative Data Analysis, 56. 115

35 descriptive codes: “drip,” “sprinkler,” or “traditional.” I applied codes for both farmers’ and fieldworkers’ explanations of irrigation use or disuse (e.g., “risk,” “distrust,” “time”). If a speaker discussed more than one of these factors in a passage, I coded for all the relevant factors. My codes are listed in Table 2.

Table 2: Coding Scheme Name Description or keywords Actors Farmer Fieldworker Community Individual with official leadership position in Leader community Irrigation type Sprinkler Drip Traditional Rain-fed, canal-fed, flood irrigation Water Discussion of infrastructure, reservoirs, canal supply maintenance, district-wide plans Explanatory factors Time Time constraint, off-farm employment, family obligation Risk Gamble, uncertainty about effect on productivity Training Don't know how to use, knowledge, human capital, extension services, similar to education code but not in formal institution Education

I conducted this analysis using NVivo software. After coding all my transcripts, I used the software to create reports by code, and used these reports to write the following two chapters. (Chapter 5 is on factors affecting

Distrust Government Personality

School, technical institute, university, similar to training code but taking place in formal institution Of NGOs, of outsiders, memory of past development failure Mayor, municipality, laws, regional or central government Based in personal or cultural characteristics, innovativeness as a personal trait, sociological explanations

Age Cost

Old age, migration of young people Financing, saving, investing, handouts, asistencialismo, matching funds

Maintenance

Disrepair, broken

farmers’ individual choices, and Chapter 6 is about the institutional context surrounding these decisions.) The purpose of this coding scheme was not an attempt to quantify my qualitative research, and I do not present frequency counts of each code. Instead, I used this scheme to discern patterns across interviewees and assemble responses into overarching insights.118                                                                                                                 118

Taylor, “Working with Data: Data Analysis in Qualitative Research (Chapter 6),” 154.

36 Supplementary Data Sources I supplemented my interview data with documents generated by El Centro Global. I examined reports El Centro Global has written about its work, both in the Lurín Valley as a whole and specifically in Lahuaytambo. I looked at previously collected survey data, but this data was too incomplete to conduct a rigorous statistical analysis. I also conducted expert interviews with Juan Sánchez, director of the Lurín Valley project; Alejandro Toledo, former president of Peru and founder of El Centro Global; Sonja Bleeker from Aquafondo, an organization that funds water resources conservation projects in Peru; and Richard Webb, former president of Peru’s Central Reserve Bank and a researcher investigating the causes of poverty in the Andes. Limitations of Study Several considerations may limit the internal validity of my findings. First, there may be interviewer bias, a common obstacle for researchers who are obvious outsiders in the communities they study.119 In addition, the presence of fieldworkers from El Centro Global at some of my interviews likely biased subjects’ responses.120 Although I began each interview by explaining the purpose of my research and my connection to El Centro Global, I had difficulty communicating my relationship with the NGO. I suspect many subjects did not fully understand how I could collaborate with the NGO on a project without being an employee. As a result, participants may have tailored their responses to what they thought El Centro Global wanted to hear. I might have lessened this bias by

                                                                                                                119

Judd, “Questionnaires and Interviews: Asking Questions Effectively,” 259. Weiss, Learning from Strangers, 144; Taylor, “Working with Data: Data Analysis in Qualitative Research,” 158. 120

37 conducting all of my fieldwork unaccompanied,121 but the logistics of travel and safety concerns made this impractical. As I have already mentioned, I used “snowballing” to find interview subjects. Consequently, my sample is not random and is biased towards individuals in the same social circle.122 A result of this bias is that many of my most forthcoming subjects are participants in El Centro Global projects. Finally, language barriers may bias my findings, albeit in a less systematic way. Although I speak good Spanish, I am not a native speaker. When speaking with interviewees I sometimes used vocabulary that was too academic or not regionally appropriate, which may have confused respondents.123 At times I had to reword a question in order to more accurately convey my meaning. Interviewees did not always hear the exact same question with the exact same wording, which may have influenced responses.124 On occasion, a coworker attempted to clarify when I had trouble expressing myself, generating further bias. I carefully designed my questions and tested them with native speakers in order to mitigate these biases. However, given the practical constraints of conducting an independent research project in a foreign country, I could not eliminate them completely.

                                                                                                                121

Weiss, Learning from Strangers, 148. Ibid, 25. 123 Judd, “Questionnaires and Interviews: Asking Questions Effectively,” 236. 124 Ibid, 247. 122

38

Chapter 5: Findings on Individual Choice The purpose of my research was to explore what motivates farmers in the Lurín Valley to adopt modern irrigation technology. I initially focused on two key factors: opportunity cost and risk perception. I include opportunity cost, perceived risk, prices, yields, installation costs, and the off-farm wage rate in the formal model of farmer decision-making that I presented in Chapter 3. I posit that farmers adopt modern irrigation technology when these factors result in an expected value of adoption that is greater than that of nonadoption: EV(adoption) = λAdopt* tAdopt*( yDry*pHigh + yRain, adopt*pLow) – cAdopt + (1-tAdopt)*w > EV(nonadoption) = λNonadopt* tNondopt* (yRain, nonadopt*pLow) + (1-tNonadopt)*w While I find some evidence supporting the opportunity cost hypothesis, I find less evidence in support of the risk perception hypothesis. Several factors that are not the focus of my hypothesis also emerged as important: installation and maintenance costs; education and training; and connection to markets.125 In this chapter, I discuss how each of these variables influences individual choice. Opportunity Cost The influence of alternate sources of income on farmer decisions is well established in the literature. Mosse notes that many farmers chose migration over farm improvements in his ethnography of a British agricultural development project in Northwest India.126 Kulecho and Weatherhead, who evaluate a drip irrigation project in

                                                                                                                125

My model includes a fixed, one-time cost and assumes farmers are integrated with agricultural commodity markets, but it does not include education and training. 126 Mosse, Cultivating Development, 68.

39 Kenya, write that although using a simple drip irrigation kit increased farmers’ incomes, “they preferred paid casual work to get the same benefit as from the technology.”127 In the Peruvian context, there are two predominant forms of off-farm employment: local nonagricultural work or migration to urban areas. Richard Webb, former President of the Central Bank of Peru and now a researcher at Universidad San Martín de Porres, noted that there has been a dramatic increase in the number of rural Peruvians who run local businesses (especially transportation services) or work as wage laborers.128 Furthermore, Webb points to migration or “oscillation” between urban and rural areas: families have a home in an urban area, where they have a formal job, and also maintain a home in their ancestral rural area. As the adoption of new technology is timeintensive, these alternative opportunities to earn income provide a disincentive to invest in drip irrigation. El Centro Global’s report on Lahuaytambo suggests relatively little nonagricultural economic activity. Only 35% of income in the district comes from nonagricultural sources.129 Producers in other areas of the Lurín Valley have developed small businesses (mainly processing fruit and dairy products), but El Centro Global reports that in Lahuaytambo “income from non-agricultural activity is also low […] and does not complement the weakness of farming activities, as occurs in other districts of the Lurín Valley.”130 However, the migration rate is very high: young people between the ages of

                                                                                                                127

Kulecho and Weatherhead, “Adoption and Experience of Low-Cost Drip Irrigation in Kenya,” 442. Webb, personal interview, September 4, 2012. 129 El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 6. 130 Ibid, 6. 128

40 25 and 49 form only 27% of the population in Lahuaytambo, compared with 35% of the population in urban areas.131 Interviews with Farmers The farmers in my sample indicate that time constraints do not permit them to investigate and install new irrigation systems. One individual who also owns a store and has a position in local government comments: Right now I don’t do it [modern irrigation] because I spend more time here in the business, in the municipality. But, if agriculture [were more important], I would apply [a new system]. Ahora no lo hago por que más estoy acá en el negocio, en la municipalidad. Casi la agricultura es poco. Por eso, sino, si tuviera la agricultura allá así, lo aplicaría.132

Many farmers recognize that El Centro Global and other NGOs working in the area provide extension services and technical assistance that would help them improve their productivity, but they say that lack of time prevents them from taking advantage of these programs. A farmer explains: Yes, they come. Various entities come here. But they don’t…like they say, lots of time. More than anything, you have to have time for that. Here we raise cattle, no? We travel by foot. They come on a Saturday, or it’s a day when they waste all your time talking. And sometimes we’re not around. Like today, it’s already 8 or 9. We’ve already left to go to our plots all day. No, we’re not around. Si vienen, acá varias entidades vienen. Pero no…como dicen, mucho tiempo. Más que todo, hay que tener tiempo para eso. Acá tenemos un ganadito, ¿no? Estamos al pie, eso… Llegan un día sábado, o es un día te quita todo el día para echar charla, eso. Y a

                                                                                                                131

Jurado, Programa De Desarrollo Ruraly y Generación De Empleo En La Cuenca Del Río Lurín, Lima: Evaluación Intermedia, 4. 132 Subject 18. December 15, 2012.

41 veces nosotros no estamos, como ahorita ya somos a las 8, 9, ya vamos a la chacra todo el día. No estamos, no.133

The migration of youth to cities is another manifestation of how the opportunity cost of farmers’ time can work against the diffusion of modern irrigation technologies. Many farmers note that young people should be the ones to experiment with new technologies, but most of Lahuaytambo’s youth had left to pursue opportunities in Lima. Most of my interviewees are older adults (see Table 3), and when asked if they had children, they often

Table 3: Age of Respondents Oldest

67

Youngest

23

Median age Mean age

50 48.8

say that they live in Lima and visit on occasion. One of my youngest interviewees says there are only “two or three” other youth residing full-time in Lahuaytambo.134 A more innovative farmer (one who is experimenting with sprinklers and drip systems) explains that many youth do not find the better opportunities they leave agriculture to pursue. He advocates training youth so that they can make agriculture more profitable and productive: We have to train the youth so that they stay in this zone, give them the ability to study and thus have more knowledge [….] so that they aren’t going to Lima because in Lima…it’s a big city where if you don’t have a profession, you die of hunger. Hay que capacitar los jóvenes para que se queden en esta zona, darles facilidades para que ellos estudien y así puedan tener mayor conocimientos [….] Y no están yendo a Lima porque en Lima…es una ciudad grande donde si no tienes una profesión, te mueres de hambre.135

                                                                                                                133

Subject 20. December 16, 2012. Subject 34. December 17, 2012. 135 Subject 37. December 19, 2012. 134

42 Farmers living in the Lurín Valley allocate their limited time between agricultural work and other activities, which prevents them from experimenting with new technologies. At the same time, the younger generation decides whether to make their livelihood in rural or urban areas. The migration of youth to Lima has eliminated most of the individuals who would be most likely to adopt new technologies. Interviews with Fieldworkers Fieldworkers report working with farmers who claim they do not have time to install new technologies or partner with NGOs. They express frustration that farmers do not see these activities as an investment of time with significant payoff: The factor of time is of utmost importance since the producers carry out multiple activities. And they think that attending a training course, or attending a talk, is, let’s say, wasting time. In reality it’s not like that. In reality it’s an investment that’s going to favor them. So I think it’s necessary to change this mentality that they have. I think it’s very necessary for them to invest their time in training themselves, in gaining knowledge, in such a way that’s going to permit them to improve the situations they currently find themselves in. El factor tiempo es de suma importancia debido a que los productores realizan múltiples actividades. Y piensan que asistir a un curso de capacitación, o asistir a una charla es, digamos, perder tiempo. En realidad no es así. En realidad es una inversión que les va a favorecer a ellos. Entonces esa mentalidad que tienen ellos, yo creo que es necesario cambiar. Pues creo que es muy necesario que ellos inviertan su tiempo en capacitarse, en conocer mas, de tal manera que les va a permitir mejorar sus actuales situaciones en que se encuentran.136

                                                                                                                136

Subject 17. December 14, 2012.

43 Fieldworkers also note that migration has robbed Lahuaytambo of those individuals who would otherwise be best positioned and most interested in adopting new technologies. One fieldworker explains: I believe it’s a function of the advanced age in many cases. In this zone…more than half the population is over 45, 50 years old. Age is a factor. […] So the youth go to the city and the older people continue getting older, and continue to have hold their property, and continue to maintain their customs of cultivation. Yo creo que es un tema de la avanzada edad de muchos casos. Que en la zona ….mas de la mitad de la población tiene mas de 45, 50 años. Es un factor de edad. […] Entonces los jóvenes se van a la ciudad y los mayores siguen haciendo viejos, y siguen teniendo sus predios, y siguen manteniendo sus costumbres de cultivo.137

Like farmers, fieldworkers recognize that the opportunity costs that farmers consider include not only the decision between various activities in rural areas but also the decision between remaining in the Lurín Valley and pursing work in urban areas. Perceived Risks Perceptions of risk have also been established as important influences on decision-making. Rogers, a sociologist known for his theory on the diffusion of innovation, writes, “The innovation carries a subjective risk to the individual.”138 Kulecho and Weatherhead write that Kenyan farmers’ perceptions of drip irrigation, which encompasses perceptions of its relative advantage, compatibility, riskiness, and complexity, affected their willingness to implement the innovation.139 Juan Sánchez Barba, project director at El Centro Global, suggests that adoption of new technologies often depends on farmers’ attitudes and perceptions. He refers to a                                                                                                                 137

Subject 17. December 14, 2012. Rogers, Diffusion of Innovations, 84. 139 Kulecho and Weatherhead, “Adoption and Experience of Low-Cost Drip Irrigation in Kenya,” 436. 138

44 vicious cycle of (little) water, (poor) technology, and (negative) mentality in the Lurín Valley.140 Following Rogers’ framework, he categorizes farmers into three types: very innovative, moderately innovative, and conservative. The smallest and most innovative group of farmers has adopted some new technologies. The second group is willing to make changes only after their more innovative neighbors have proved that they are advantageous. The final group, which also contains the poorest farmers, resists change. According to Sánchez, the most innovative farmers are willing to “gamble on” drip irrigation, while the others perceive it as too risky. These risks include uncertain access to water, seeds, markets for outputs, and technical assistance, which are also factors that El Centro Global can help provide. The NGO, which needs to show that its projects generate results, likewise has to “gamble” on the few farmers willing to adopt, which can explain why a few farmers in a community reap disproportionate benefits.141 Interviews with Farmers I hypothesized that many interviewees would explain that they have not installed drip or sprinkler irrigation because they perceive the technology to be risky or are uncertain about the benefits. As predicted, some nonadopters explain that they do not know whether modern irrigation systems improve productivity because they have not yet tested them: No, as we aren’t used [to modern irrigation], maybe it can also turn out well. We still don’t use it here. Nobody uses it here. There are some people who are testing [it] on their farms…this is how we can find out how it results.

                                                                                                                140 141

Sánchez Barba, personal interview. August 29, 2012. Ibid.

45 No, como no estamos acostumbrados, de repente también puede resultar bien. Todavía acá no usamos, pues. Todavía acá casi nadie usa. Hay algunas personas que están probando en sus chacras…así se puede saber como les da.142

At the same time, many nonadopters voice the opinion that sprinkler or drip irrigation was the less risky proposition, and offer substantive explanations as to why it is an improvement over the traditional system: It saves water. It saves time. It saves everything. Se ahorra el agua. Se ahora el tiempo. Todo se ahorra.143 Disadvantage? No. Rather, it benefits the plant. There isn’t…with drip irrigation nothing gets spoiled. [The water goes directly to] the plant…there’s a drip, drip, drip. It [adjusts to the amount the plant needs]. The plant stays green. So I don’t think…there’s no disadvantage. ¿Desventaja? No. Más bien favorece el cultivo. No hay en que….por que con goteo no se malogra nada. Se fija en la plantita…hay gota, gota, gota. Va tomando la medida de la planta. Se mantiene verde. De ahí no creo que…no hay nada de desventaja.144

In sum, I found mixed evidence regarding my hypothesis that the perception of drip and sprinkler systems as risky technologies prevents farmers from using them. Many farmers have not observed the benefits of these systems with their own eyes, and thus are uncertain about their effect on productivity. However, as many farmers who believe modern irrigation is technically superior fail to adopt, perception of risk is not a central explanatory factor. Interviews with Fieldworkers Fieldworkers characterize farmers’ varying perceptions and level of aversion to risk as a personality trait. One interviewee explains:                                                                                                                 142

Subject 23. December 16, 2012. Subject 29. December 16, 2012. 144 Subject 25. December 16, 2012. 143

46 They’re producers, some of them, who are averse, reluctant to take risks, and others who are more willing to take on proposals that mean work for them, but also the possibility of a better future. Son productores, algunos de ellos, que son aversos, renuentes a tomar riesgos, y otros que si son mas dispuestos a asumir propuestas que les impliquen a ellos trabajo, pero también posibilidades de un mejor futuro.145

The fieldworker adds that farmers are not just wary of the risks implied by changes in their irrigation method, but of other proposals El Centro Global presents: They approach [the idea] with a lot of caution because it is something new, obviously. A lot of the time there is incredulity at the beginning, which is also natural. And this disappears when they can really perceive a benefit from changing their approach. But as I tell you, at times these processes are not very quick. La toman con mucha cautela por que es algo nuevo, obviamente. Hay muchas veces incredibilidad al principio, lo que es natural, también. Y eso se va a revertir cuando vean que realmente se puede percibir un beneficio cambiando la modalidad. Pero como te digo esos procesos a veces no son tan rápidos.146

Fieldworkers perceive farmers as generally skeptical of new technologies, which contrasts with the mixed opinions I found among farmers. Simply put, the fieldworkers I spoke with view overcoming farmers’ suspicion as a component of their work. Installation and Maintenance Costs The upfront and ongoing costs associated with sprinklers and drip irrigation play a major role in whether a rational farmer choses to use the new systems. Anticipating that costs would be a major driver of the adoption decision, I attempted to select a field site where this was less of an issue because irrigation systems had previously been installed.                                                                                                                 145 146

Subject 38. December 19, 2012. Subject 38. December 19, 2012.

47 Several interviewees confirmed that the local municipality and the comunidades campesinas working in the area had installed sprinkler systems on communal lands between 2002 and 2004.147 When I arrived to conduct interviews, respondents explained that the systems had gone unused. After many years of disuse, costly repairs are necessary before they can use the systems. An all-inclusive drip irrigation kit (with seeds as well as hoses and pumps) costs about 8000 soles (3075 USD)148 per hectare for peas and 10,000 soles (3844 USD) per hectare for alfalfa.149 These cost vary with the size of the parcel converted to drip irrigation, so a farmer experimenting on a quarter-hectare parcel would need 2000-2500 soles (769-961 USD) for installation. Nevertheless, farmers in Lahuaytambo are unlikely to be able to afford modern irrigation technology without significant savings, access to credit, or government subsidies. Lahuaytambo lacks financial services, which precludes credit access and complicates the already difficult task of saving money.150 The comunidades favor using their limited funds to build canals and reservoirs, and the municipality cannot legally fund projects with purely private uses.151 Interviews with Farmers Farmers report that high costs are the main reason they have not installed modern irrigation systems: Because…money. The problem is the money. If I had money, I would install [modern irrigation].

                                                                                                                147

This project was sponsored by FONCODES (Fondo de Cooperación al Desarrollo Social) before a change in law banned using public funds for projects with private uses. Sánchez Barba, personal interview. December 20, 2012. 148 One New Peruvian Sol = 0.3844 USD as of May 11, 2013. 149 Sánchez Barba, personal interview. December 20, 2012. 150 Abhijit V. Banerjee, Poor Economics, 192. 151 Former President Toledo says that Peruvian law allows for financing “up to the farm gate” (hasta la puerta de la chacra). Toledo, personal interview. January 17, 2013.

48 Por que… money. El problema es el money. Si hubiera, si lo instalaría.152

Farmers often mention that they would be interested in installing new systems if there were government or NGO support: [We ask] that they support us with modern irrigation. Modern irrigation with tubes, sprinklers, that is, everything that is necessary for modern irrigation, everything. Because it’s costly. We, individually, we can’t take on this cost. [Pedimos] que, sea, que nos apoye con riego tecnificado. Riego tecnificado con tuberías, aspersores, o sea, todo lo que se necesita para riego tecnificado, todo. Por que es costoso. Nosotros, individualmente, no podemos asumir ese gasto.153 Cleary [modern irrigation] is good. As I say, if there were support from somebody, maybe we could have modern irrigation. But, no, we can’t because we don’t have sufficient funds to buy tubes. Claro que es bueno. Como te digo, si hubiera una ayuda de alguien, de repente nosotros podemos tener el tecnificado. Pero no, no podemos por que no tenemos suficiente para poder comprar tubos.154

The few farmers who have adopted modern irrigation systems did so either through partnerships with NGOs or from their own savings. One farmer explains: [I financed it] with my own resources. I have mainly homemade sprinklers, made by a relative who [knows how to build modern irrigation systems]. Con mis propios recursos. Yo tengo mayormente aspersores de casero, preparados por un familiar que está en este riego tecnificado.155

The same farmer brags about the return on his investment: For each packet of alfalfa they pay you 25 soles in Lima. If you take an average of 30 packets, 100 packets, we can calculate how much money you’re going to earn. So…and this is each month. […] For example, this year, from April to August, I earned

                                                                                                                152

Subject 34. December 17, 2012. Subject 20. December 16, 2012. 154 Subject 21. December 16, 2012. 155 Subject 37. December 19, 2012. 153

49 more from my farm than from my pension [as a retired teacher]. So, with this I can live a little better, or I can achieve something in my fields. Our community lacks this vision. Por cada paquete de alfalfa en Lima te pagan 25 soles. Si llevas un promedio de 30 paquetes, 100 paquetes, saquemos la cuenta de cuánto dinero va a dar. Entonces…y eso es mensual. […] Por ejemplo, yo este año, de abril a agosto, sacaba ganancias de mi chacra más que mi sueldo [como profesor jubilado]. Entonces, ya con esto puedo vivir algo, un poquito mejor, o puedo lograr algo en mis parcelas. Esta visión falta nuestra comunidad.156

Installing the tubes, pumps, hoses, sprinkler heads, and other equipment necessary for drip or sprinkler irrigation is a considerable cost for farmers in the Lurín Valley. However, the few farmers who have managed to finance this investment report a considerable return. Interviews with Fieldworkers Fieldworkers acknowledge that the costs of installing irrigation systems are not insignificant but remark that it is possible to save for these expenses: Yes, it’s very difficult [to save money], but I think if one decides to do it it’s possible. We can see in the case of [name omitted], he’s a person who has decided, and well, he’s the first to put modern irrigation in his field. This required an investment. Si es muy difícil [ahorrar dinero], pero yo pienso que si uno se decide si puede. Vamos a ver en el caso de [nombre omitido], es una persona que se ha decidido, y bueno, es el primer a hacer riego tecnificado en su parcela. Para eso se necesita inversión.157

Fieldworkers complain about a handout culture among rural Peruvians and blame the government of Alberto Fujimori in the 1990s for creating excessive expectations of government welfare:                                                                                                                 156 157

Subject 37. December 19, 2012. Subject 30. December 17, 2012.

50 There’s an issue among our rural population: for many years we had “asistencialista” [pro-welfare] governments, governments that gave gifts. Specifically, Alberto Fujimori’s government. It was a decade when the community’s cultural patterns were broken. That is to say, the communal spirit works [by organizing community members], but when he began to give gifts, to give materials [….] So people got used to receiving things when a [development] project arrives. Hay un tema que en nuestra población rural: durante muchos años hemos tenido gobiernos asistencialistas, gobiernos que regalaban muchas cosas. Concretamente de Alberto Fujimori. Fue una década donde se fracturaron los patrones culturales de las comunidades. Es decir el espíritu comunal llega a conseguir todo vía organización, pero cuando él empezó a regalar cosas, a dar activos [….] Entonces la gente se ha acostumbrado a que le den algo cuando llega algún proyecto.158

Fieldworkers from El Centro Global proudly state that their project does not give handouts but instead focuses on extension services and technical support. However, one fieldworker expresses the view that the project could be more effective by providing matching funds to farmers’ investments. He explains that El Centro Global should finance most, but not all, of an investment, and the farmers should provide the rest: We should have some kind of…offering, let’s say in some sort of inputs or goods, but […] conditioned on some financing on their part. And it seems to me that it’s important to have this. Debemos tener alguna…algún ofrecimiento, vamos a decir, en algún tipo de insumos o bienes, pero […] condicionado a la entrega de una contrapartida por parte de ellos. Y me parece que también es importante contar con eso.159

                                                                                                                158 159

Subject 17. December 14, 2012. Subject 38. December 19, 2012.

51 The fieldworkers I interviewed are aware of the considerable cost of modern irrigation technology, but are also wary of breeding dependence on aid. They discount the difficulty of saving money in a zone without financial services. Education and Training Education, training, and human capital are important determinants of adoption of technology. Education or training not only allows individuals to evaluate the benefits of new technologies, but also lessens the learning curve associated with new technologies or techniques. Rogers writes that greater mental ability and human capital is a characteristic of early adopters.160 In evaluating the adoption of drip irrigation kits in Kenya, Kulecho and Weatherhead find that most farmers had little formal education or agricultural training, which affected their ability to assess the drip irrigation project.161 Schultz, who argues that traditional agricultural systems are efficient with the limited factors of production they posses, emphasizes that the key to improving traditional agriculture is through improving human capital.162 The residents of Lahuaytambo have low levels of human capital. An El Centro Global report finds that only 49% of Lahuaytambo farmers have some secondary education. 163 My sample was better educated than average: 56% had completed secondary school and 22% had some form of post-secondary education (see Table 4). Lahuaytambo has a primary and secondary school, but students must go to Lima for higher education. For two years, Lahuaytambo had a state-funded agricultural institute to

                                                                                                                160

Rogers, Diffusion of Innovations, 189. Kulecho and Weatherhead, “Adoption and Experience of Low-Cost Drip Irrigation in Kenya,” 438. 162 Schultz, Transforming Traditional Agriculture, 175. 163 El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 9. 161

52 train local youth in farming practices, but it closed in 2011 due to low enrollment.164 The high rate of child malnutrition (48%)165 in the district may also contribute to stunting, reduced cognitive ability, and lower levels of educational attainment.166 El Centro Global calls the low level of human capital “a limitation to the introduction of […] more

Table 4: Educational Attainment* Less than secondary school

4

22%

Finished secondary school

10

56%

4

22%

More than secondary school *Only 18 of 23 interviewees were asked this question

advanced technologies.”167 Interviews with Farmers Farmers point to a lack of technical training and know-how as an important reason why they have not used existing modern irrigation or sought to install new systems. One farmer explains: Because the community of Pariapongo, which has already installed sprinklers for irrigation, it has installed more or less three hectares…four hectares more or less [….] But they don’t use it for a lack of knowledge, and technical support as well. Por que la comunidad Pariapongo, que tiene instalado ya para riego por aspersión, tiene instalado mas o menos tres hectáreas…cuatro hectáreas más o menos [….] Pero no lo utiliza por falta de conocimiento, de apoyo técnico también.168

Another farmer describes how a lack of training and support has prevented him from installing new irrigation systems: Because…no…there’s no…nobody trained us, nobody gave us [materials], nobody taught us or gave us these…these…the hoses. That’s it.

                                                                                                                164

Sánchez Barba, “Conversation with Juan Sánchez Barba,” December 20, 2012. El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 4. 166 Abhijit V. Banerjee, Poor Economics, 32. 167 El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 9. 168 Subject 18. December 15, 2012. 165

53 Por que…no no no…no hay este…quien nos capacitó, quien nos dio, quien nos enseñe o quien nos traiga los, este…todo este…las mangueras. Eso es.169

The individuals who are active participants in El Centro Global programs explain that they participate for the training the NGO provides: I know have to—what’s it called?—train myself…nobody is born knowing. You have to learn. It’s clear, logical. Sé que tengo que, como se llama, capacitarme…nadie nace sabiendo. Hay que aprender, claro, lógico.170

The farmers I spoke with recognize the importance of training. Some have received this training from El Centro Global or other NGOs, but others feel they lack opportunities to develop these skills. Interviews with Fieldworkers While many farmers complain that there are few opportunities to receive training in Lahuaytambo, fieldworkers seek to fill this gap. One fieldworker explains how they use the approach of field demonstrations to teach farmers a range of agricultural skills: We use participatory methodologies based in learning by doing, basically demonstrative, a methodology that allows for experimenting in the field […] Basically it’s a matter of doing, repeating, experimenting. Nosotros usamos metodologías participativas […] basadas en aprender al hacer, básicamente demostrativas, una metodología que permite experimentar en campo[….] Básicamente es un tema de hacer, replicar, experimentar.171

Another fieldworker talks about the challenges of attracting participants to these training programs. Those who attend regularly want to learn, but she says of the others:

                                                                                                                169

Subject 28. December 16, 2012. Subject 12. December 12, 2012. 171 Subject 17. December 14, 2012. 170

54 In contrast to the others, we’ve invited them to a ton of events and they don’t come [….] There’s no interest in getting out ahead, or…how do you say it? That is, they want to stay there. They don’t want to get out. For all that we say, “We’re going to do this.” “Ah, I want to stay, I’m already old, my dear,” is what some of them say, “You teach this to the young people. Al contrario de las otras personas, les hemos invitado a un montón de eventos y no llegan. [….] No hay interés por salir adelante, o como decirlo, o sea, quieren quedar allí. No quieren salir. Todo lo que decimos de, "Vamos a hacer esto." "Ah quiero quedar, ya estoy viejo mi amor," es lo que dicen algunos, "Eso se enseña a los jóvenes."172

The fieldworker continues, explaining why she is committed to this difficult work: For me, mainly it’s to teach that yes, it’s possible, that it’s not only a matter of waiting for someone, or for NGOs to come, or for someone else from the state to come. In my own field, [I want] to teach them how it’s done. So that they can see, or at least have the opportunity to ask, “Why do you do it like that and can’t I do it as well?” Yo, mayormente es enseñar que si se puede, que no es solo esperar de alguien, o que venga ONGs, o que venga otro del estado. Bueno, no, o si ellos dicen que no que no, enseñarle a ellos. En el propio campo que yo tengo, enseñarles a como se hace. Para que ellos vean, o al menos tienen una posibilidad de preguntar, “¿Por que tu haces así y yo no puedo hacer eso?”173

The desire to share their knowledge with Peruvian farmers motivates many of the fieldworkers I interviewed. Accordingly, the mixed success they have had engaging participants is a source of frustration. Market Integration Adopting modern irrigation systems, a new technology that allows producers to both increase their yields and sell at high off-season prices, increases incomes only if                                                                                                                 172 173

Subject 34. December 17, 2012. Subject 34. December 17, 2012.

55 Table 5: Market Integration and Crops

farmers are connected to the market. Access to agricultural commodity markets was a fundamental assumption in my model, but 57% of my interview subjects

No connection to market

13

57%

Potato

17

74%

Wheat

14

61%

Broad beans

13

57%

Peas

12

52%

5

22%

Most common crops

Median number of crops 174

report that they do not grow for sale on the market (see Table 5).

The most common

crops are starchy staples that they consume at home. Without a market connection, farmers do not care to produce a quantity greater than the amount they can consume, and price signals do not have any incentive. In assuming that farmers have access to markets, I have made a mistake that Sen alleges is common in economic research: considering only market-based mechanisms for acquiring food.175 In Lahuaytambo, as in many rural areas of the developing world, production for home consumption remains significant. Juan Sánchez, the director of El Centro Global’s Lurín Valley project, emphasizes the importance of connecting farmers to the market, which allows productivity gains to result in additional income.176 In fact, a large part of the NGO’s work consists of organizing farmers into groups that collectively produce a quantity large enough to take to market. Sánchez stresses that the project’s value is in bringing together multiple components: enhancing productivity with modern irrigation, growing high-value crops like aguaymanto (golden berry), organizing farmers into groups, and connecting them with buyers in Lima.177

                                                                                                                174

The proportion of farmers growing for market is relatively low, but the proportion of total quantity produced reaching the market is higher. That is, a relatively small number of farmers grow a majority of the region’s produce and sell it to markets in Lima. El Centro Global para el Desarrollo y la Democracia, Diagnóstico del Distrito de Lahuaytambo, 8. 175 Sen, “Food, Economics, and Entitlements,” 194. 176 Sánchez Barba, “Conversation with Juan Sánchez Barba,” December 20, 2012. 177 Ibid.

56 Interviews with Farmers Nonadopters are most likely to produce only for personal consumption. One farmer explains: Yes, it’s only for here, to take home. No, no, no, I don’t have any for business. When [farms] are for business, they’re larger, then. Normally we plant for home consumption, nothing more. Si, solamente es para acá, para llevar a casa. No, no, no tengo para negocio. Ya, cuando sirven para negocio, son mas grandes, pues. Normalmente sembramos para consumo de casa, no mas.178

Another farmer, one who has implemented some innovations and sells to the market, talks about how making a living in agriculture requires integrating with markets: They don’t have that vision of the future, because mainly they grow for personal consumption. But, then, we have to see that…we have to grow for personal consumption and for sale. No tienen esa visión del futuro, por que mayormente ellos cultivan para autoconsumo. Pero, entonces, nosotros tenemos que ver que… hay que cultivar para autoconsumo y para venta.179

This interviewee emphasizes the importance of not only selling to the market but also offering high-quality, high-value crops: People always grow potatoes, wheat, barley…but these products can’t compete with those that come, for example, from Argentina. So we have to grow other products. Siempre la gente cultiva papa, trigo, cebada…pero esos productos no compiten con los que vienen, por ejemplo, de Argentina. […] Entonces hay que implementar otros productos.180

                                                                                                                178

Subject 33. December 17, 2012. Subject 37. December 19, 2012. 180 Subject 37. December 19. 2012. 179

57 In addition to adopting new irrigation technologies and growing high-value crops, these farmers are also more likely to belong to an association of producers that pools produce from many plots to sell in Lima. A farmer explains his involvement: If each person cultivates […] and from that we bring together a quantity and go down to the farmers’ market. We have three farmers’ markets right now in Lima, every Saturday. Each farmer grows his crops…potatoes, beans, and he who has…he who’s going to the market takes the products. It’s by rotation. For example, it’s my turn for six months, then someone else for six months. Si cada uno cultiva […] y de ello reunimos la cantidad y baja a la feria. Tenemos tres ferias ahorita a Lima, todos los sábados. Todos cultivan sus productos…papas, alverjas, y él que tiene…él que está yendo a feria lleva el producto. Es por rotación. Por ejemplo, a mi me toca seis meses, después otra persona seis meses.181

The farmers in my sample recognize that adopting new technologies is most beneficial as part of a broader strategy to integrate with Peru’s market for agricultural commodities. Some farmers have embraced this strategy, but others have hesitated. Interviews with Fieldworkers The fieldworkers note that many farmers resist the growing crops for the market: But here in the high part [of the valley] it’s a little very difficult. Here people, as you must have realized, mainly what they produce is for their own consumption. Not for sale. And they’ve resigned themselves to this. Pero acá en la parte alta es un poco muy difícil. Acá la gente como te habrás dado cuenta mayormente lo que produce es para el autoconsumo de ellos. No para venta. Y se conforman con eso.182

                                                                                                                181 182

Subject 19. December 16, 2012. Subject 37. December 19, 2012.

58 One of the fieldworkers I interviewed specializes in helping farmers form associations, and another specializes in commercialization. They discuss the difficulty of persuading farmers to join together to sell their products: And for what? So that the small surpluses that they can obtain are commercialized in a joint manner. That is to say, “I can offer this man 50 kilos, 40 kilos,” which is very little, but if we organize ourselves, we could have a more considerable quantity that we can offer to the market. ¿Y para qué? Para que los pequeños excedentes que pueden obtener sean comercializados de manera conjunta. Es decir, "Yo puedo ofrecer este señor 50 kilos, 40 kilos," que es algo muy poco, sino que juntándonos, podamos tener un volumen mas considerable que se puede ofrecer ya al mercado.183

Although fieldworkers view El Centro Global’s strategy of improving agricultural productivity and integrating with markets as the best path toward improving rural incomes and generating sustainable economic growth, they recognize and understand farmers’ resistance to this market-based approach. Discussion In this chapter, I have presented several factors that enter a rational farmer’s decision to adopt modern irrigation technologies: opportunity cost, perception of risk, maintenance and installation costs, education and training, and connection to markets. In some cases, I find agreement between farmers and field workers as to which factors most influence decisions: both farmers and fieldworkers reported that limited time prevents farmers from experimenting with new technologies. Both groups of respondents also agreed on the technological benefits of drip or sprinkler irrigation systems. Importantly,

                                                                                                                183

Subject 38. December 19, 2012.

59 both farmers and fieldworkers recognized that adoption of modern irrigation technologies makes most sense for farmers growing for profit. I also encountered points of disagreement. Farmers complained that they had inadequate training, while fieldworkers expressed their frustration at finding few participants for their training programs. Furthermore, farmers insisted that the cost of modern irrigation system was prohibitive while fieldworkers claimed that farmers could save more money if they decided to do so. Despite this disagreement, I believe that the high cost of irrigation technology is one of the most pressing barriers to adoption. Not all of these findings fit into the model I developed before fieldwork. My model incorporates opportunity cost, perception of risk, and costs. Market integration is an underlying assumption. It does not adequately incorporate education and training or a farmer’s decision to enter the market. Furthermore, my individual treatment of each of these characteristics does not capture how the factors interact to shape farmers decisions. First, I have presented training and time as two separate factors. I model a farmer as dividing his time between farm and nonfarm activities, but I do not explicitly include training time. In this chapter, I first discuss farmers’ time constraints, and I later introduce lack of training as an additional constraint. However, training is a constraint because it requires time. Acquiring the knowledge necessary to install and use drip or sprinkler systems is an investment. In the short run, a farmer may have to give up another activity (for instance, spending time with family or working as a wage laborer) to attend an NGO-sponsored workshop, but in the long run the skills he acquires will allow him to farm the same plot of land more efficiently, which could give him more time to devote to other activities.

60 In addition, many of Lahuaytambo’s youth (those arguably in the best position to take advantage of agricultural training) have migrated to Lima and therefore do not participate in NGO-sponsored workshops. Youth must leave Lahuaytambo to pursue vocational training or post-secondary studies, which creates an additional reason to migrate. Individuals who receive post-secondary education are likely to have higher offfarm earning potential, so they are unlikely to return to agriculture. Finally, the interaction between market integration and risk perception is key. I present market integration and risk as two separate factors, a result of incomplete reasoning while planning my research. I focused on the risks surrounding the new technology rather than the potentially larger risks brought by integration into the market economy. The qualitative methodology that I use aims not only to assess whether farmers perceive risk but also to identify sources of risk. I find relatively little evidence supporting the idea the drip or sprinkler irrigation is are risky technologies. However, following my fieldwork, I realized that the source of risk may not be the technology itself, but the fact that transitioning from subsistence to market agriculture, using the technology to increase production, brings additional volatility and risk. The risk of market integration is well established in the literature. In Prosperity and Violence, Bates writes that industrial societies have markets for risk but that agrarian societies do not. Farmers such as the individuals I interviewed in the Lurín Valley “must directly bear the costs of uncertainty; they must self-insure.”184 Farmers transitioning from traditional to market agriculture leave behind these self-insurance mechanisms but do not immediately acquire market- or government-provided insurance. Adopting modern irrigation technologies means growing for profit, rather than subsistence, and exposes                                                                                                                 184

Bates, Prosperity and Violence, 38.

61 farmers to the volatility of agricultural markets. The risk of market integration is a component of farmers’ decisions that I would ask farmers to discuss explicitly if I were to do this project again. The individual choice framework presented in this chapter has allowed me to systematically examine the effects of each factor on an individual farmers decisionmaking and compare interview data from farmers and fieldworkers. I have identified points of agreement between the two groups of respondents—on the benefits of sprinkler systems and the limited time available to farmers—and points of disagreement about which constraints were more binding—most notably, regarding the costs of adoption. This approach has also prompted me to think about how these factors interact. In the next chapter, I explore the institutional context in which these farmers’ make their decisions and reevaluate whether my individual choice model is an appropriate framework to apply to these decisions.

62

Chapter 6: Institutional Context and Collective Choice Up to this point, I have focused on the factors that might influence a rational actor in his individual choice to adopt drip or sprinkler irrigation technology. However, while conducting my fieldwork, I noted that many of my respondents spoke of the adoption decision as a collective, rather than individual decision. A unique local institution, the comunidad campesina, shapes the lives of rural Peruvian farmers. The comunidades hold communal lands, manage water resources, and make other communal decisions. I found that I could not understand farmers’ adoption decisions without considering the role of the comunidades campesinas. In this chapter, I discuss how these local institutions shape the adoption decisions of my subjects in the Lurín Valley. I give background on the history and role of the comunidades. Then I present evidence from my interviews as to how the comunidad affects farmers’ decisions. I conclude by discussing the interplay of individual and collective decision making in the case of the Lurín Valley, Peru. Evolution of the Comunidades Peru’s comunidades campesinas have evolved from the strong kinship ties that have marked Andean society since pre-Inca times. Individuals belonged to a lineage called an ayllu, which traced decent from a common ancestor. Additionally, members had claims “to productive resources including the support and labor of one another, on the basis of the relationship established between them and defined by themselves and their culture as kinship.”185 The Inca incorporated these units into their empire, which allowed

                                                                                                                185

Spalding, Huarochirí, 48.

63 them to mobilize labor for considerable construction and agriculture projects.186 The Spanish conquerors also took advantage of these pre-existing structures to facilitate organizing the labor force and collecting taxes.187 Although Andean society evolved in response to the Inca Empire, the Spanish Conquest, and Peru’s independence, the communal approach to economic activities and decision-making has endured. The ayllu, and later the comunidad campesina, collectively managed factors of production such as land, labor, and inputs. For instance, members were obliged to harvest one household’s fields with the expectation that the ayllu would do the same for every other household in turn. The Spanish viceroyalty formed reducciones de indígenas, or indigenous settlements, around the ayllus.188 These indigenous groups later evolved into the comunidades campesinas. Although the exact form of the comunidad campesina varies from place to place, most comunidades have a similar governing structure and set of responsibilities. Today, the Peruvian constitution and numerous Peruvian laws codify the role of the indigenous comunidades campesinas. 189 The state recognizes them as autonomous juridical persons with the authority to allocate lands among members, determine communal production, and require participation in communal labor. 190 Qualified commoners (members over 18 years of age who have fulfilled communal labor obligations) can participate in the community’s General Assembly, which, led by an elected president, makes these decisions.191 Figueroa explains that “the comunidad is not a simple aggregation of                                                                                                                 186

Ibid, 50. Ibid, 121. 188 Cotlear, Desarrollo Campesino En Los Andes. 40. 189 Inter-American Commission on Human Rights, Chapter X: The Rights of Indigenous Communities, section B, paragraph 13. 190 Ibid. 191 Congress of the Peruvian Republic, Ley General de Comunidades Campesinas. 187

64 families, but a social medium where certain economic relationships between members are established and where collective decisions and activities take place.”192 Although some scholars believe the land reform of the 1970s, which expropriated and redistributed 38% of the country’s agricultural land, 193 actually hurt the ability of the comunidades to organize their members, the comunidades remain dominant institutions in Andean society.194 Cotlear writes that the comunidades play an important role in the spread of new technologies. He explains, “The diffusion of appropriate technologies can be facilitated or impeded by the presence of economic institutions […] The comunidades campesinas regulate many of the family’s production decisions. An important topic […] is the relationship between technological change and the transformation of the community.”195 Like Cotlear and Figueroa, I find that I cannot examine the decisions of Andean farmers without considering the influence of the comunidad campesina. Evidence Interviews with Farmers I spoke with individuals belonging to all three of the comunidades campesinas in Lahuaytambo (see Table 6).196 These interviews reveal the extent to

Table 6: Membership in Comunidades Campesinas Llambilla

10

Pariapongo

6

Santa Ana

1

Non member

5

which farmers viewed the adoption of modern irrigation technologies as a collective, rather than individual, decision. When asked if they had plans to install new equipment, many community members say they would do so if the community did:                                                                                                                 192

Figueroa, La Economía Campesina En La Sierra Del Perú, 2. USAID, Property Rights and Resource Governance Country Profile: Peru, 4. 194 Cotlear, Desarrollo Campesino En Los Andes, 76. 195 Ibid, 25. 196 I was unable to acquire data about enrollment in each of the comunidades, so I do not know how my sample compares with the population. I suspect that the community of Llambilla is overrepresented, and I am certain that Santa Ana (whose members live further from the district’s main town) is underrepresented. 193

65 [I want] everyone, the whole town, to unite to do a single…a single irrigation [project], to get a better product. [Quiero] que todos, todo el pueblo, se unan para hacer una sola…un solo riego, para sacar un producto mejor.197 We still don’t have, as they say, the abilities to install [irrigation]. It would be a part, we would have to convene the community so that it would be productive [….] From what I can tell, yes, yes we would like to possibly achieve this some day. I hope that some day we can. Todavía no tenemos por acá como dice todas las facilidades para instalar. Sería una parte, tendríamos que convocar la comunidad par que sea productivo [….] Por lo que veo si, si deseamos que en un día posiblemente realizar eso. [….] Ojalá que un día pueda.198

That subject, who is optimistic about potential communal action to improve irrigation systems in Lahuaytambo, expresses powerlessness to make individual changes. After explaining some of the shortfalls of the current flood irrigation system, she says: If it’s not sufficient, what are we going to do? It’s tradition…. Si no es suficiente, ¿qué vamos a hacer? Es costumbre.…199

As I have already mentioned, the comunidad campesina has responsibility for allocating who will work communal lands, coordinating public works, and maintaining public goods such as canals and reservoirs. My research indicates that some individuals transfer this collective choice orientation to the decision to install drip or sprinkler equipment. This equipment is divisible and excludable: an individual could choose to install it on the plots he or she cultivates without the cooperation of his neighbors and with a low risk of expropriation. A farmer could install a new irrigation system on his                                                                                                                 197

Subject 19. December 16, 2012. Subject 31. December 17, 2012. 199 Ibid. 198

66 farm and continue to collaborate with the comunidad to secure water access. Nevertheless, the fact that the comunidades campesinas have traditionally managed the area’s water resources precluded many of my interview subjects from viewing the adoption decision as an individual choice. Individuals who do not belong to a comunidad campesina (23% of my sample) are generally excluded from the reservoir and canal system. These interviewees explain that joining the community (and fulfilling the communal works required of members) is the only way to secure access to water: “The problem is that I’m not a member of the comunidad. Water is only for members, so those who are not members, even if they have land, can’t water. They don’t allow them water.” El problema es que no soy comunero. El agua es solamente para comuneros, entonces los que no son comuneros, aun si tengan chacra, no pueden regar. No les permiten el agua.200

For the most part, individuals who do not belong to the comunidades have not sought out other methods to secure water access. Instead, they rely entirely on seasonal rains. Without access to communities’ (albeit insufficient) reservoirs, they have no way to supply water to a more efficient modern irrigation technologies.  The subjects I spoke with seem resigned to low farm productivity and an outsider status in Lahuaytambo. My fieldwork shows that the comunidades campesinas govern many aspects of water resources management in the Lurín Valley. They not only take charge of public goods, such as canals and reservoirs, that serve the entire community, but set the standard for what techniques and technologies are used in individual plots. For many farmers, the                                                                                                                 200

Subject 35. December 18, 2012.

67 decision to adopt drip or sprinkler irrigation technologies depends in large part on what is the norm in the comunidad. Interviews with Local Leaders A few of my interviewees hold prominent positions in Lahuaytambo, either as leaders of the district municipality or their respective comunidad campesinas. I spoke with the current mayor and a former mayor of Lahuaytambo. The mayor serves four-year terms (without term limits) and must consider the interests of the three different comunidades in Lahuaytambo: Pariapongo, Llambilla, and Santa Ana.201 Both respondents emphasize the importance of collaborating with the comunidades: We [the municipality] see the necessity of having communities. So, what does the municipality do? What does it have to do for this community? It has to build a dam or improve its canals as well. The same for the other communities. Dams, canals, roads, whatever it may be. All of this…we are bound together. We’re bound together, we work in coordination with the presidents of the communities. Nosotros vemos la necesidad que tiene las comunidades. Entonces, ¿qué hace la municipalidad? […] ¿Que hay que hacer para esta comunidad? Hay que hacer una represa [...] o hay que mejorar sus canales también. […] De igual manera para las otras comunidades. Represas, canales, sus caminos, sea. Todo eso…estamos ligados. Estamos ligados, trabajamos en coordinación con los presidentes de las comunidades.202

Similarly, the former mayor comments: Working as mayor means working with the communities. The rural municipality is…it’s not its role to be improving the streets, improving something else in the urban part but going to the fields, improving the dams, the canals, training people in modern irrigation, introducing other products that aren’t…aren’t traditional in Lahuaytambo.

                                                                                                                201 202

Subject 26. December 16, 2012. Ibid.

68 Trabajar como alcalde es trabajar con las comunidades. […] El municipio rural es…no es su autoridad estar mejorando las calles, mejorar otra cosa en la parte urbana sino ir al campo, mejorar las represas, los canales, capacitar a la gente en riego tecnificado, implantar otros productos que no…que no son tradicionales en Lahuaytambo.203

That respondent speaks about how, as mayor, he led the installation of sprinkler systems on lands held by the communities, but, as noted elsewhere in this thesis, these systems were not used: I installed, in my community an average of 50 hectares of modern irrigation. In another community an average of two hectares…I’m mistaken, 20 hectares. But unfortunately the people, for lack of training, haven’t used it. They prefer irrigation by gravity.” Instalé en mi comunidad, en Pariapongo, un promedio de 50 hectáreas de riego tecnificado. En otra comunidad un promedio de dos hectáreas…me equivoqué de 20 hectáreas. Pero desgraciadamente la gente por falta de capacitación, no lo ha utilizado. Ellos prefieren el riego a gravedad.204

An individual who has served as president of one of the comunidades also speaks of working with the municipality, but complains that these collaborations are rarely fruitful: As president I have the obligation to do works, projects with the district council or the region of Lima, the provincial mayor, other entities. But almost the majority of what I propose they don’t allow because…you know they delay two years, three years, sometimes they delay four years…. Como presidente tengo obligación de hacer trabajos, proyectos con el consejo distrital o con la región de Lima, alcaldesa provincia, otras entidades. Pero casi la mayoría de que

                                                                                                                203 204

Subject 35. December 19, 2012. Subject 35. December 19, 2012.

69 me presenté no me dieron curso por que…sabe que demoran dos años, tres años, a veces demoran cuatro años….205

In sum, community and municipality leaders have focused their efforts relating to water resources on water supply (by building dams and canals). Only one interviewee attempted to improve water distribution through the installation of drip and sprinkler irrigation. Officials in the municipality and leaders in the comunidades campesinas recognize the importance of working with each other, but have so far had mixed results. Interviews with Fieldworkers In contrast with officials in the local municipality, fieldworkers at El Centro Global have little experience working with the comunidades campesinas. They often work with individual farmers who are active community members, but they have made few attempts to involve entire communities in their projects promoting modern irrigation or other innovations: With the peasant community in reality we have worked very little, very little. At the beginning a small gathering, and after, nothing more. Con las comunidades campesinas en realidad hemos trabajado muy poco, muy poco. Al principio un pequeño acercamiento, y después, ya no fue más.206

The NGO has sought to form farmers into groups of six to ten (much smaller than the comunidades, which can have several hundred members) who can pool their surplus produce and share the task of taking it to market.207 They then work with these groups on improving various aspects of their production and business—planting high value crops, applying fertilizer, getting organic certifications, tracking revenue and expenses, and of course, installing modern irrigation systems.                                                                                                                 205

Subject 25. December 16, 2012. Subject 38. December 19, 2012. 207 Subject 17. December 14, 2012. 206

70 El Centro Global publications recognize the important role of the comunidades in rural districts: a recent report calls the communities the “institutions with the greatest power to promote innovations.”208 Despite this rhetoric, the organization has not focused on incorporating the comunidades into its activities. The same report comments that the comunidades are among “a group of actors that still doesn’t have a significant presence in the program.”209 El Centro Global has not fully engaged this local institution that plays an important role in shaping the perceptions and decisions of farmers in the Lurín Valley. Discussion In this chapter, I have shown that local institutions—in my case study, the comunidades campesinas—exert an influence on the diffusion of technology. This analysis contrasts with the previous chapter, where I demonstrated that a rational choice framework relates many of the factors that enter a farmer’s decision. I now turn to the question of how to reconcile the aspects collective and individual choice. Although I cannot definitively say whether an individual or communal orientation dominates the decisions of farmers in Lahuaytambo, I can discuss the conditions where one orientation toward decision-making seems stronger than the other. Rogers, in Diffusion of Innovation, explains that some individuals make decisions regardless of what the rest of their social system does, others individuals require an intermediate level of approval, and a third group only makes a change when the entire social system does so as well.210 Very few farmers in my sample of respondents had begun to experiment with modern irrigation technologies, but those who had seem to fit                                                                                                                 208

Jurado, Programa De Desarrollo Rural y Generación De Empleo En La Cuenca Del Río Lurín, Lima: Evaluación Intermedia, 30. 209 Ibid, 32. 210 Rogers, Diffusion of Innovations, 15.

71 my rational choice model. These farmers are also better educated and grow a substantial portion of their crop for the market. They fall into the first group Rogers describes: they consciously make a decision that breaks with the status quo, presumably because they calculate that it is beneficial. In contrast, my rational choice model may not explain the choices of individuals who require their community to approve and promote any innovation they may adopt. Even if adopting modern irrigation technologies would bring positive net benefits to these farmers, they will not do so because it is not the community norm. The power of social norms precludes these individuals from undertaking the type of cost benefit analysis that governs the decisions of rational actors. In my sample, these farmers are also less educated and grow almost entirely for home consumption. Most farmers likely fall in between these two extremes: the costs and benefits of adoption as well as prevailing social norms in making their decisions. Nevertheless, my research supports the theory that community institutions matter for the diffusion of innovations. El Centro Global has had some success in promoting drip and sprinkler irrigation among farmers who do not rely on prevailing norms to dictate their decisions and therefore fit into my rational choice framework. However, they have been less successful in spreading these technologies among farmers who adhere more closely to the status quo. I do not reject my hypothesis that rational factors—opportunity cost, perceptions of risk, knowledge and training, maintenance costs, and market integration— influence farmers’ decisions, but I recognize that the strength of the comunidades campesinas may preclude some farmers from undertaking this calculation. Accordingly,

72 El Centro Global, and all other actors attempting to promote new technologies, cannot do so on a large scale without engaging existing local institutions.

73

Chapter 7: Conclusion In the previous chapters I have discussed how the adoption of modern irrigation technologies is at times an individual decision, at times a collective decision, and at times a hybrid decision for farmers in Lahuaytambo. In this final chapter, I present some policy implications of these findings and discuss the generalizability of my results. I conclude with suggestions for further research related to the diffusion of technology and agricultural development. Policy Recommendations My work has led to a number of policy recommendations for El Centro Global and actors in the Peruvian government. Some of these recommendations may also be useful to state actors and NGOs working in agricultural development in other regions and countries. These various actors, who all aim to improve the lives of the rural poor, should seek to integrate smallholder farmers with markets and pay special attention to the provision of financial services. They should also make sure that the technologies they promote are appropriate for the context they are working in, and engage existing community structures to promote innovations. Integrate Smallholders with Markets First, El Centro Global should continue and heighten their efforts to connect farmers with formal markets. I found that a relatively small number of farmers (43% of my sample) grew crops for sale. Helping farmers reach markets is a key component of El Centro Global’s overall strategy for rural development. According to Juan Sánchez, the director of El Centro Global’s Lurín Valley program, the organization tries to assist farmers with every step of production: planting high-value crops like papa native

74 (indigenous potato varieties), improving fertilizer and irrigation practices, organizing farmers into groups that can bring a sizable surplus to market, and teaching farmers to advertise and manage their costs. 211 This valuable work should continue. Local, regional, and national government actors also need to play a role in connecting farmers to markets. Most notably, the Lurín Valley and other rural areas of Peru need improved transportation infrastructure. Some residents already make the sixplus hour journey to Lima every week to sell produce or visit family members. Widening and paving mountain roads would reduce travel time, increase safety, and allow more farmers to make the journey. The government has spearheaded an “explosion” in road building in recent years, 212 and should continue these efforts. Another contribution government actors could make is to subsidize Internet and mobile phone service in the Lurín Valley. Many of my interview subjects have cell phones and email accounts that they use when they travel to Lima. Providing these services in Lahuaytambo would allow farmers to communicate with buyers and obtain information about market prices. Provide Financial Services Furthermore, for farmers in the Lurín Valley to become successful entrepreneurs, they need access to financial services. Many of my interview subjects pointed to the high cost of irrigation equipment as the main barrier to adoption. There are currently no financial services—microfinance or traditional banks—operating in Lahuaytambo and other rural areas of the Lurín Valley.213 Farmers need access to credit and savings instruments in order to invest in irrigation technology and other farm projects. According to Sánchez, the El Centro Global is seeking to collaborate with banks to offer small in                                                                                                                 211

Sánchez Barba, “Conversation with Juan Sánchez Barba,” August 29, 2012. Webb, personal interview, September 4, 2012. 213 Sánchez Barba, personal interview. December 20, 2012. 212

75 kind loans.214 This “seed bank” (banco de semillas) is a start, but state and civil society actors could help provide much-needed services to a group traditionally overlooked by the commercial banking sector. In addition, farmers in the Lurín Valley need another financial instrument: insurance. My work, along with previous studies, highlights the uncertainty inherent in traditional agriculture.215 Without formal insurance, farmers self-insure by making conservative decisions, but this strategy prevents them from integrating with the market economy and benefitting from new technologies.216 Banerjee and Duflo write that microinsurance is an untapped market: “The poor face an enormous amount of risk and should be willing to pay a reasonable insurance premium [….] With billions of poor people waiting to be insured, even a tiny profit per policy could make it a tremendous business proposition.”217 Although some of the fieldworkers I spoke to doubt that rural farmers, who are often suspicious of financial institutions, would make use of financial services, I believe that access to credit, savings, and insurance could make a dramatic difference in the lives of Lurín Valley farmers. Introduce Appropriate Technologies Another recommendation to NGOs and other actors in agricultural development is to pay attention to whether they are introducing a technology that is appropriate for their context and whether they introduce it at an appropriate pace. In the case of Lahuaytambo, I believe the drip and sprinkler technologies El Centro Global advocates is appropriate. Farmers using modern irrigation in nearby districts, most notably San Andrés de                                                                                                                 214

Sánchez Barba, “Conversation with Juan Sánchez Barba,” August 29, 2012. Abhijit V. Banerjee, Poor Economics, 142; Bates, Prosperity and Violence, 38. 216 Bates, Prosperity and Violence, 38. 217 Abhijit V. Banerjee, Poor Economics, 147. 215

76 Tupicocha, have had considerable success, and early innovators in Lahuaytambo report positive outcomes as well.218 However, both fieldworkers at El Centro Global and I recognize that it is not appropriate for farmers to immediately convert all of their fields to modern irrigation. Figueroa writes that “modernization makes more sense when there is a mix of modern inputs with improved traditional inputs,” and the farmers I spoke with intuitively understand this point.219 Those who have begun to use drip and sprinkler systems are experimenting with the new technology on only a portion of their fields. Not every innovation is an appropriate innovation for a given set of circumstances. Likewise, immediate and complete adoption is not necessarily the best adoption process. Collaborate with Existing Institutions Finally, and perhaps most importantly, I recommend that El Centro Global and other NGOs work with existing community structures—in this case, the comunidad campesina—to generate large-scale adoption of new technologies. The comunidades exert significant influence on their members. Therefore, convincing community leaders of the advantages of modern irrigation may result in the entire community more promptly adopting the technology. When Figueroa complains, “Neither technological development nor programs for technological diffusion are thought of in terms of the rural peasant economy,”220 he means that state actors and NGOs do not leverage pre-existing institutions to achieve social progress and economic growth. El Centro Global has realized some success working with individual farmers with an entrepreneurial streak and a desire to take advantage of what the organization has to offer. I believe the NGO can

                                                                                                                218

Sánchez Barba, “Conversation with Juan Sánchez Barba,” August 29, 2012. Figueroa, La Economía Campesina En La Sierra Del Perú, 137. 220 Ibid, 138. 219

77 complement this approach with efforts to engage the comunidades campesinas in the Lurín Valley. Generalizability I conducted my field research in a small, 837-person district in the Peruvian Andes. This site, like any other site I could have chosen, has its idiosyncrasies. In particular, Lahuaytambo differs from other areas of the sierra in that it is very close to Lima. Many residents travel between the metropolis and their rural home every few weeks. Furthermore, the indigenous Quechua language, which is still the first language for Peruvians in more isolated parts of the sierra, is extinct in Lahuaytambo and the Lurín Valley. Despite these differences, Lahuaytambo’s climate, geography, and prices of factors such as land and labor are similar to those in other areas of the Andes. As a result, my study of El Centro Global’s efforts to promote modern irrigation systems is relevant to actors working in agricultural development in these regions. Even more broadly, my examination of the appropriateness of a rational choice framework may inform situations that do not closely resemble my area of study. In the next two sections, I discuss how my findings might generalize outside Peru and outside the agricultural sector. Outside Peru Farmers around the world face similar production decisions. They must choose a course of action that will allow them to fulfill multiple (and at times contradictory) goals: maximizing income, minimizing risk, ensuring food security, and protecting the environment. I believe my research shows that a rational choice model is a valuable framework for organizing the factors that enter farmers’ decisions. A change agent

78 seeking to promote a new agricultural technology would do well to consider (and if possible, quantify), the costs (both monetary and time) of an innovation, the level of human capital required to implement it, how it will affect farmers’ yields, whether farmers are integrated with markets that will allow them to capitalize on increased production, and how risky farmers perceive the technology to be. At the same time, my findings show the importance of looking beyond standard economic models. My individual choice model did not capture the influence of traditional collective institutions in farmers’ decision-making. As many economists, political scientists, and sociologists before me have noted, institutions shape humans’ economic, political, and social behavior. The comunidades campesinas have historically made important economic decisions in Andean communities, and their influence continues to this day. In any country, organizations seeking to effect change should look to how they can collaborate, rather than compete, with existing local institutions. Outside Agriculture Although agriculture is a unique economic activity in many respects, some of my findings may apply to development projects in other sectors. My finding with respect to the importance of institutions is particularly robust. Institutions are equally important outside of the agricultural sector, if not more so. As countries develop, they depend even more on unwritten rules of the game to guarantee safe and mutually beneficial economic, political, and social activities. Actors trying to speed along the process of development and modernization should pay attention to the institutions that already exist and the institutions they are trying to shape. Sometimes existing institutions may oppose change

79 in order to protect their interests, but in other cases change agents may leverage them to achieve their goals. Further Research and Final Thoughts In this thesis I have presented the results of a year-and-a-half-long research project. I set out to identify the factors that enter farmers’ decisions regarding the adoption of modern irrigation technologies, and I sought to answer this research question using a qualitative methodology involving interviews with farmers and fieldworkers. My two research trips to Peru generated some evidence in support of two factors that I expected to be important: the opportunity cost of implementing new technologies and the perception of the technology as risky. I found more convincing evidence indicating that training, costs, and connection to market are important considerations; these factors easily fit into my rational choice framework. However, my fieldwork brought to my attention the role of Peru’s comunidades campesinas play in making economic decisions, and this influence does not fit into my individual rational choice model. My research has raised many questions and prompted me to think about how they might be answered. I would make many changes if I were to carry out this project again. Agricultural development is a field of research that benefits from experimentation, and furthermore, one where experimentation is possible. NGOs like El Centro Global are able to pilot different programs. They try a new intervention, and if it works, other civil society and state actors can replicate it. Carefully designing and monitoring these programs allows researchers to quantify the effects of different approaches. In the realm of irrigation technologies, I would like to see research that

80 experiments with different levels of subsidies for the new technologies and different training programs. Economic development often means transitioning from a closed economy and limited worldview to greater integration at the regional, national, and global levels. This process is inherently risky and uncomfortable. During my fieldwork I met many individuals who wanted to secure a better future for themselves and their families but still approached many of the decisions they might make (installing drip irrigation, leaving the farm for the city, growing a high-value crop, opening a small store) with trepidation. My findings underscore the need for research as to what instruments business, government, and civil society can deploy to smooth the process of economic growth and integration.

81

Works Cited Abhijit V. Banerjee. Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. 1st ed. New York: Public Affairs, 2011. Alfaro Moreno, Julio César, and Ricardo Claverias Huerse. Diagnóstico Socioeconómico de La Cuenca Del Río Lurín. Lima: Centro Global para el Desarrollo y la Democracia y Centro de Investigación, Educación y Desarrollo, July 2010. Aquafondo. Gestión Eficiente del Riego Tecnificado Para Cultivos Rentables. Aquafondo, October 2011. Bates, Robert H. Prosperity and Violence: The Political Economy of Development. 1st ed. New York: W. W. Norton, 2001. Bertrand, Marianne, Sendhil Mullainathan, and Eldar Shafir. “A Behavioral-Economics View of Poverty.” The American Economic Review 94, no. 2 (May 1, 2004): 419– 423. Burney, Jennifer A., and Rosamond L. Naylor. “Smallholder Irrigation as a Poverty Alleviation Tool in Sub-Saharan Africa.” World Development 40, no. 1 (January 2012): 110–123. “CGDD firma acuerdo con el BID para ejecutar un proyecto en el valle de Lurin.” El Centro Global Website, October 29, 2009. http://cgdd.org/es/programas/proyvalle. Congress of the Peruvian Republic. Ley General de Comunidades Campesinas, 1987. http://faolex.fao.org/docs/pdf/per20093.pdf. Conway, Gordon. The Doubly Green Revolution: Food for All in the Twenty-First Century. Ithaca, N.Y.: Comstock Pub. Associates, 1998. Cotlear, D. Desarrollo Campesino En Los Andes: Cambio Tecnológico y Transformación Social En Las Comunidades de La Sierra Del Perú. 1a ed. Lima [Peru]: Instituto de Estudios Peruanos, 1989. Dhawan, B. D. “Drip Irrigation: Evaluating Returns.” Economic and Political Weekly 35, no. 42 (October 14, 2000): 3775–3780. Duflo, Esther. “Poor but Rational?” In Understanding Poverty, by Abhijit Banerjee, Roland Benabou, and Dilip Mookherjee, 366–378. New York: Oxford University Press, 2006. http://economics.mit.edu/files/747. Duflo, Esther, Michael Kremer, and James Robinson. “Nudging Farmers to Use Fertilizer: Theory and Experimental Evidence from Kenya.” American Economic Review 101, no. 6 (October 2011): 2350–2390. El Centro Global para el Desarrollo y la Democracia. Diagnóstico del Distrito de Lahuaytambo. El Centro Global para el Desarrollo y la Democracia, October 2012. Felipe-Morales, Carmen. Diagnóstico participativo sobre la gestión del agua en la cuenca del Río Lurín. El Centro Global para el Desarrollo y la Democracia, October 2010. Figueroa, Adolfo. La Economía Campesina En La Sierra Del Perú. Lima: Pontificia Universidad Católica del Perú, Fondo Editorial, 1981. Food and Agriculture Organization. FAOStat. Food and Agriculture Organization of the United Nations. Accessed October 24, 2012. http://faostat.fao.org/site/339/default.aspx.

82 ———. Perfíl de País: Indicadores de Seguridad Alimentaria. Food and Agriculture Organization of the United Nations, November 2011. http://www.fao.org/fileadmin/templates/ess/documents/food_security_statistics/co untry_profiles/esp/Peru_S.pdf. Inter-American Commission on Human Rights. Chapter X: The Rights of Indigenous Communities. Second Report on the Situation of Human Rights in Peru. InterAmerican Commission on Human Rights, June 2, 2000. www.cidh.oas.org/countryrep/Peru2000en/chapter10.htm. Judd, Charles M., Eliot R. Smith, and Louise Kidder. “Questionnaires and Interviews: Asking Questions Effectively.” In Research Methods in Social Relations, 228– 265. 6th ed. Fort Worth: Holt, Rinehart, and Winston, 1991. Jurado, Alejandro Rafael. Programa de Desarrollo Rural y Generación de Empleo En La Cuenca Del Río Lurín, Lima: Evaluación Intermedia. Banco Interamericano de Desarrollo, November 2012. Kulecho, I.K., and E.K. Weatherhead. “Adoption and Experience of Low-Cost Drip Irrigation in Kenya.” Irrigation and Drainage 55 (2006): 435–444. Meinzen-Dick, Ruth. “Beyond Panaceas in Water Institutions.” Proceedings of the National Academy of Sciences of the United States of America 104, no. 39 (September 25, 2007): 15200–15205. Miles, Matthew B. Qualitative Data Analysis: A Sourcebook of New Methods. Beverly Hills: Sage Publications, c1984. Mosse, David. Cultivating Development: An Ethnography of Aid Policy and Practice. London: Pluto Press, 2005. North, Douglass C. “Institutional Change and Economic Growth.” The Journal of Economic History 31, no. 1 (March 1, 1971): 118–125. ———. “Institutions.” The Journal of Economic Perspectives 5, no. 1 (January 1, 1991): 97–112. Ostrom, Elinor. “Collective Action and the Evolution of Social Norms.” The Journal of Economic Perspectives 14, no. 3 (July 1, 2000): 137–158. Quiroga, Eduardo R. “Irrigation Planning to Transform Subsistence Agriculture: Lessons from El Salvador.” Human Ecology 12, no. 2 (June 1, 1984): 183–201. Rogers, Everett M. Diffusion of Innovations. New York: Free Press of Glencoe, 1962. Sachs, Jeffrey. The End of Poverty: Economic Possibilities for Our Time. New York: Penguin Press, 2005. Schultz, Theodore William. Transforming Traditional Agriculture. Chicago: University of Chicago Press, 1983. Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press, 1998. Sen, Amartya. “Food, Economics, and Entitlements.” In Agricultural Development in the Third World, 189–205. edited by Carl Eicher and John Staatz, 1990. Sivertsen, Arne, and Anders Lundberg. “Farming Practices and Environmental Problems in an Arid Landscape: A Case Study from the Region of Lambayeque, Peru.” Geografiska Annaler. Series B, Human Geography 78, no. 3 (January 1, 1996): 147–161. Spalding, Karen. Huarochirí  : an Andean Society Under Inca and Spanish Rule. Stanford, California: Stanford University Press, 1984.

83 Tax, Sol. El Capitalismo Del Centavo: Una Economia Indígena de Guatemala. Guatemala: Centro Editorial Jose de Pineda Ibarra, 1964. Taylor, Steven J., and Robert Bogdan. “Working with Data: Data Analysis in Qualitative Research.” In Introduction to Qualitative Research Methods: A Guidebook and Resource, 134–163. 3rd ed. New York: Wiley, 1998. Thurlow, James, Samuel Morley, and Alejandro Nin Pratt. “Lagging Regions and Development Strategies: The Case of Peru.” IFPRI Discussion Paper Series (September 2009). http://www.ifpri.cgiar.org/sites/default/files/publications/ifpridp00898.pdf. Timmer, C. Peter. A World Without Agriculture: The Structural Transformation in Historical Perspective. Washington, D.C.: AEI Press, 2009. Tversky, Amos, and Daniel Kahneman. “Rational Choice and the Framing of Decisions.” The Journal of Business 59, no. 4 (October 1, 1986): S251–S278. USAID. Property Rights and Resource Governance Country Profile: Peru. USAID, December 2010. http://usaidlandtenure.net/sites/default/files/country-profiles/fullreports/USAID_Land_Tenure_Peru_Profile.pdf. Weiss, Robert Stuart. Learning from Strangers: The Art and Method of Qualitative Interview Studies. 1st Free Press pbk. ed. New York: Free Press, 1995. World Bank. World Data Bank: World Development Indicators and Global Development Finance. World Bank, September 28, 2012. http://databank.worldbank.org/data/home.aspx.

84

Appendix I: Interview Questions English Translation Farmer Interviews Introduction (prepared script) “I am an American student from Stanford University. I am conducting a research project on the factors that influence adoption of drip irrigation technology and relations between farmers and NGOs like El Centro Global. Over the next few days, I will be interviewing farmers such as yourself throughout Huarochirí’s Lurín Valley. Participation in these interviews is voluntary. My research will be kept confidential, meaning I will not use your name to identify your response. I’ve brought this recorder with me so that I can record your exact response and pay attention while you talk. Can I turn it on? “I have 15 questions to ask you, and the interview will take approximately 10-15 minutes. I will begin by asking you for some background information. Then I will ask you to discuss the irrigation techniques you use and describe your participation in NGO projects. There are no right or wrong answers. I appreciate your time. Are you ready to begin?”

Questions on farm background and method of production 1. What is your name, place of residence, comunidad campesina, and age? 2. How many years of education have you completed? 3. How many members are in your family? (Adults/children?) 4. How many plots do you have and how large are they? 5. What crops do you grow? 6. What fraction of your output is for personal consumption? 7. When and where do you sell your output? 8. Do you hold any off-farm jobs do you hold? a. What are they? b. How much time to they take? Questions on participation in NGO projects 1. How do you irrigate your fields? a. What other options do you have? 2. What is your opinion of other irrigation techniques? a. Drip irrigation? b. Sprinklers? c. Flood irrigation? 3. Have you discussed your irrigation method with fieldworkers from El Centro Global or another NGO? a. What do they recommend? 4. Please describe your interactions with fieldworkers from El Centro Global. a. How often do you speak with fieldworkers? b. What events have you attended? c. Drip irrigation project? Another project? d. When? How long? 5. (If respondent described significant participation in a project) Why did you choose to participate in this project?

85

6.

7. 8. 9.

a. Perceived benefits? (If respondent did not participate) Why didn’t you choose to participate in the project El Centro Global was promoting? a. Perceived risk? b. Other time commitments? (If respondent participated) What were the advantages of participating in this project? (If respondent participated) What were the drawbacks of participating in this project? Please describe the experience of your friends, family members, or neighbors who participated in El Centro Global’s projects. a. Anyone I should talk to? Can you introduce me?

Fieldworker Interviews Introduction (prepared script) “As you already know, I am a student from Stanford University in the United States. I am conducting a research project on the factors that influence adoption of drip irrigation technology, using the irrigation projects conducted by El Centro Global para el Desarrollo y la Democracia here in the Lurín Valley as a case study. I have spent the past few days interviewing farmers in Lahuaytambo, and now I would like to get the perspective of fieldworkers such as yourself. Participation in this interview is voluntary, and I predict no harm or benefit form participating in an interview. My research will be kept confidential, meaning I will not use interviewees’ names to identify their responses.” “I have 15 questions to ask you, and the interview will take approximately 30-45 minutes. I will begin by asking you for some general information on El Centro Global’s project in Lurín. Then I will ask you to discuss your perception of how farmers make decisions and your personal experience as a fieldworker. There are no right or wrong answers. I appreciate your time. Are you ready to begin?”

Questions on project background 1. Please describe the history and evolution of the irrigation project? 2. What are the primary goals of this technical irrigation project? 3. What is project timeline? Questions on perceptions of farmer decision-making 1. Please tell me about participation in the technical irrigation project. a. Barriers to participation? b. Assets, selling points? 2. How do you describe the project to potential participants? 3. What arguments or strategies do you use to encourage farmers to participate? a. Situations when farmers shouldn’t adopt? 4. When farmers choose not to participate, what reasons do they give for their decision? a. What do you think of their explanations? 5. What sets apart the individuals who choose to work with you from those who do not? 6. Please tell me about one or two past projects you’ve worked on. a. Participation?

86 Questions on personal experience 1. (First and last name) 2. How long have you been working for this particular organization? 3. How long have you been working in the area of agricultural development? 4. What is your educational background? 5. What motivates you to work in agricultural development? 6. How would you describe your role as a fieldworker? 7. In your opinion, what can your organization (or NGOs in general) do to improve their impact on the wellbeing of the target community? Versión en castellano Entrevistas con producores Introduction (preparada) “Soy una estudiante estadounidense de la Universidad Stanford. Estoy llevando a cabo un estudio sobre el uso de riego tecnificado y las relaciones entre productores y ONGs como El Centro Global. En los días que vienen, entrevistaré varios agricultores como usted. Participar en una entrevista es voluntario. Mi estudio es confidencial; es decir, no usaré su nombres para identificar su respuesta. He traído conmigo esta grabadora para que pueda grabar sus respuestas y prestar atención cuando habla. ¿Puedo encenderla? “Tengo 15 preguntas y la entrevista tomará aproximadamente 10-15 minutos. Empezamos con algunos datos básicos. Luego, pediré que describa su uso de riego y sus colaboraciones con El Centro Global. No hay respuestas correctas ni incorrectas. Le agradezco su tiempo. ¿Listo para empezar?”

Preguntas básicas 1. ¿Cómo se llama? ¿Dónde vive? ¿Cuál es su comunidad campesina? ¿Qué edad tiene? 2. ¿Cuántos años de educación cumplió? 3. ¿Cuántos miembros hay en su familia? ¿Cuántos son adultos y cuántos son niños? 4. ¿Cuántos terrenos tiene y de que tamaño son? 5. ¿Qué cultiva en cada chacra? 6. ¿Qué parte de sus cultivos es para autoconsumo? 7. ¿Dónde vende sus cultivos y con qué frecuencia? 8. ¿Usted tiene un empleo que no tiene que ver con la agricultura? a. ¿Qué es? b. ¿Cuánto tiempo toma? Preguntas sobre el uso de riego y participación en proyectos ONG 1. ¿Qué tipo de riego usa en sus chacras? a. ¿Cuáles otras opciones tiene? 2. ¿Qué opinas de otros tipos de riego? a. Riego por goteo b. Aspersores c. Riego por gravedad o inundación 3. ¿Usted ha discutido su método de riego con un trabajador del Centro Global? a. ¿Qué recomienda?

87 4. Por favor, descríbame sus interacciones con trabajadores del Centro Global. a. ¿Con qué frecuencia habla con trabajadores del Centro Global? b. ¿A cuáles eventos ha asistido? c. Proyecto de inundación por goteo, otro proyecto d. ¿Cuándo? ¿Por cuánto tiempo? 5. (Si el participó en un proyecto) ¿Por qué eligió participar en este proyecto? a. Beneficios 6. (Si no participó) ¿Por qué no eligió participar? a. Riesgos percibidos b. Otras obligaciones 7. (Si participó) ¿Cuáles son las ventajas de participar? 8. (Si participó) ¿Cuáles son las desventajas? 9. ¿Conoce algunos amigos, relaciónes, o vecinos que participaron en proyectos del Centro Global? Por favor, descríbame sus experiencias. a. ¿Hay alguien con quién debo hablar? ¿Podría introducirme? Entrevistas con trabajadores de campo Introducción (preparada) “Como usted ya sabe, soy estudiante de la Universidad Stanford en los Estados Unidos. Estoy llevando a cabo un estudio sobre el uso de riego tecnificado y las relaciones entre agricultores y ONGs. Uso los proyectos de riego del Centro Global para el Desarrollo y la Democracia acá en la provincia de Huarochirí como un estudio de caso.” “Acabo de entrevistar varios agricultores en la cuenca Lurín y ahora quiero hablar con trabajadores de campo como usted. Participar en esta entrevista es voluntario, y participar no lleva algún daño ni beneficio. Mi estudio es confidencial; es decir, no usaré los nombres de los entrevistados para identificar sus respuestas.” “Tengo 15 preguntas y la entrevista tomará aproximadamente xx minutos. Empezamos con información básica sobre el proyecto del Centro Global. Luego, pediré que describa su percepción de las decisiones de los agricultores y sus experiencias personales en el campo. uso de riego y sus colaboraciones con El Centro Global. No hay respuestas correctas ni incorrectas. Le agradezco su tiempo. ¿Listo para empezar?”

Preguntas sobre el proyecto en general 1. ¿Podría discutir el tema de participación en el proyecto de riego tecnificado? a. ¿Obstáculos a atraer participantes? b. ¿Ventajas? 2. ¿Cómo describe el método de riego por goteo a los agricultores que puedan adaptarlo? 3. ¿Cuáles argumentos o estrategias usa la ONG para animar los campesinos a usar el riego por goteo? a. ¿Hay situaciones en las cuáles no deben usarlo? 4. Cuando los agricultores eligen no usar riego tecnificado, ¿cómo le explican su decisión? a. ¿Qué piensa usted de sus explicaciones? 5. ¿Cuáles atributos diferencian los agricultores que eligen participar en el proyecto de los que eligen no participar?

88 6. ¿Podría describirme uno o dos proyectos en los cuales trabajó en el pasado? a. Participación Preguntas sobre experiencia personal 1. (Nombre y apellido) 2. ¿Cuánto tiempo lleva con el Centro Global? 3. ¿Cuánto tiempo lleva trabajando en el campo de desarrollo agrícola? 4. ¿Cuál es su formación? 5. ¿Cuál es su motivación para trabajar en el campo de desarrollo agrícola? 6. ¿Cómo describiría su papel como trabajador de campo? 7. ¿En su opinión, cómo podría su organización mejorar su impacto en el bienestar de la comunidad?

89

Appendix II: Photos   The Lurín Valley is located in the coastal Andes, just a few hours from the metropolis of Lima. June 2011.

Approaching Lahuaytambo on the unpaved road from Antioquía. The town is located at approximately 3300 meters (10,800 feet). September 2012.

Water supplies build up in small reservoirs such as this one during the rainy season (January-March). December 2012.

90 Members of the comunidad campesina of Llambilla meet each Sunday to determine who gets access to limited water supplies each week. December 2012.

An El Centro Global worker teaches farmers about fertilizer techniques. July 2011.

These buses provide transportation between Lima, Lahuaytambo, and other towns in the Lurín Valley. Buses run 3-4 days each week and the trip can take 6-7 hours. December 2012.

91 Anna with a bag of maíz morado, or purple corn. September 2012.

One of the few alfalfa fields in Lahuaytambo with a sprinkler irrigation system. September 2012.

Anna and a fieldworker from El Centro Global. September 2012.