CHAPTER 6 MANA: Improving Food and Nutrition Security in Antioquia, Colombia JAMES GARRETT The Food and Nutrition Improvement Plan of Antioquia (Plan de Mejo- ramiento Alimentario y Nutricional de Antioquia, or MANA) began in 2001 as an outgrowth of one Colombian politician’s passion to reduce child mortality from malnutrition. Probably very few people in Colombia’s mountainous department of Antioquia know what the letters of MANA stand for. But thanks to successful, energetic promotional campaigns among poli- ticians, communities, and other stakeholders and an effective synthesizing of new and existing programs across multiple sectors, quite a few people know MANA. This chapter explores the research questions and hypotheses described in Chapter 4 in the context of this program. The Department of Antioquia Antioquia is one of the departments of Colombia (Figure 6.1). It is located in the central northwestern part of the country but has a narrow strip that reaches to the Caribbean. Most of its territory consists of high mountains and steep valleys. The department covers 62,840 square kilometers and has an esti- mated population of about 5.7 million, according to a 2005 census. The depart- ment is probably better known for its capital, Medellín. Medellín is the second largest city in Colombia, with about 2.2 million people in the municipality and 3.7 million in the metropolitan area. Although known for its mountains, Antio- quia’s geography represents a broad range of climate and landscapes, including seashore, plains, lakes, rivers, swamps, forests, and jungles (Colombia 2005; Gobernación de Antioquia 2006; Alcaldía de Medellín 2006). The mountainous terrain has resulted in a degree of developmental isola- tion for Antioquia. It was affected less than other parts of the country by the armed conflict that wracked Colombia during the nineteenth century, and instead of the estates found in much of the rest of the country, smaller farms 100 with relatively fewer financial and natural resources predominated. These circumstances gave rise to an entrepreneurial, industrial spirit, and in time significant intraregional trade and commerce developed (Wikipedia 2009). Mining, agriculture, and manufacturing were the mainstays of the Antio- quian economy during the nineteenth century and most of the twentieth (Wikipedia 2009). Agriculture, mining, and manufacturing are still important, but services now account for at least 60 percent of the economy. Information technology and medical services are well developed. Antioquia now provides 25 percent of Colombia’s nontraditional exports and 15 percent of its gross domestic product (GDP) (Colombia 2004; Gobernación de Antioquia 2006). Current political institutions arise, for the most part, from the 1991 Con- stitution. Although not a federal government, Colombia is decentralized in many respects. For instance, more than 40 percent of spending is allocated by subnational governments (Alesin 2005). Colombia’s departments and munici- palities have less autonomy than Brazilian states or Argentine provinces, but IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 101 COLOMBIA • 62,840 square kilometers • Varied terrain, largely mountainous • 5.7 million population • 3.7 million in metropolitan Medellín ANTIOQUIA Medellín Figure 6.1 Political and topographical maps of Antioquia, Colombia Sources: GADM (2010); Jarvis et al. (2008). the constitution and the electoral process confer on them substantial political legitimacy. In addition, decades of guerrilla warfare have led to de facto, if not constitutional, autonomy in large sections of Colombia, including parts of Antioquia (Dillinger and Webb 2001). Colombia weathered severe economic and political crises from the late 1970s to the mid-1990s. Severe economic recession; the rise of guerrilla groups, particularly the Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia, or FARC); and the emergence of the cocaine trade in the 1980s turned the cities, suburbs, and countryside in Antioquia into battlefields for narcotraffickers, the military, police, and paramilitary and guerrilla groups (Hylton 2006; Wikipedia 2009). During the next two decades, Antioquia experienced intense violence, and an estimated 40,000 of Medellín’s population between the ages of 14 and 25 died violently (Hylton 2002). The assassination of prominent narcotrafficker Pablo Escobar in 1993 marked the beginning of the decline of the Medellín Cartel and of violence in Antioquia (Kurtz-Phelan 2007; Colombia 2009). The situation has improved, but the FARC and other armed groups still operate in the department, and social and economic development still lags (Colombia 2009; U.S. Department of State 2009). Compared to other depart- ments in Colombia, Antioquia is poor. According to the 2005 census, Antio- quia’s overall poverty rate, as measured by Unsatisfied Basic Needs, is 22.6 percent, compared to only 9.2 percent in the Department of Bogotá. The poverty burden is even higher in rural areas, at 47.1 percent as opposed to 15.4 percent in urban areas (Colombia 2005). Methods and Data for the Case Study For two weeks in June 2006, I met in Antioquia with a variety of stakeholders. I interviewed more than 60 people affiliated with MANA. These individuals had worked with MANA at different points in its history (from conception to the present) and had played various roles in the program (including conceptu- alizing, managing, implementing, and observing it). I conducted focus-group interviews with the implementing partners of each of the six programmatic components of MANA. These discussions usually lasted between one and two hours. These groups included universities, regional development associations, pro- fessional associations, and private-sector firms. I also conducted two focus- group interviews in the field with the counterpart municipal staff who were responsible for implementing the program at the municipal level, one in a rural area and another in an urban community. In addition, I held meetings with a regional implementing team. This group was composed of the different operational partners at the regional level, each of whom is responsible for 102 CHAPTER 6 implementing a different component. This coordinating group is responsible for working together among themselves as well as with the municipalities to make sure that MANA is implemented in a coordinated way. I also met with the consulting firm responsible for, on the demand side, strengthening the municipalities’ abilities to fulfill their obligations to the program (particularly planning and management responsibilities) and with representatives from national programs with whom MANA coordinates on some activities. MANA staff set up the focus groups but invited representatives of all implementing organizations affiliated with the component. MANA staff did not participate in the meetings. I separately held formal meetings with the program director and the senior managers of MANA and had informal conver- sations with staff responsible for overseeing specific components. In addition, I conducted extensive key informant interviews with the principal actors, that is, those individuals who had started MANA, those who managed it, and those policymakers who played a role in its initiation and implementation. These individuals provided in-depth histories of the ini- tiation and development of MANA. Key informants included the director of MANA; principal staff from the Ministry of Health who were responsible for initiating MANA; partner staff at the Ministry of Agriculture and Ministry of Education at the time MANA was being developed; the minister of planning and the governor’s special adviser at the time of the creation of MANA; the current governor of Antioquia and relevant ministers (Health, Education, Agriculture, and Planning); and a legislator whose committee provided over- sight of the program. The interviews used open-ended questions, largely following the guide shown in Table 4.1, although these were adapted to the appropriate time period for each group or individual. I took extensive notes, distinguishing between summaries of the conversations and verbatim records. In almost all cases, the conversations were also tape recorded. Given the written record, the interviews were not transcribed, but the tapes could be referred to as needed during analysis. I then consolidated the written responses to the interviews. This effort was fairly straightforward as I had followed the outline provided in Table 4.1 and had asked each focus group similar questions. With those who had been with MANA from the start, I pursued the entire chronology. No quantitative software was used. I consolidated and analyzed all interview notes, paying attention to the highlights and explanations proffered by informants but also relating responses to the components of the conceptual model and hypotheses. PowerPoint presentations, government documents, communications mate- rials, and other studies and documentation on MANA complemented these interviews. Although there was no specific search for noncorroborative evi- IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 103 dence, I tried to interview all sides of the partnership from different perspec- tives over time. As part of MANA’s learning process, implementing organiza- tions and community partners had prepared presentations and documents on lessons learned that proved especially useful. Getting Started In 2000, Guillermo Gaviria of the Partido Liberal political party was elected governor of Antioquia. He was soon recognized as one of the outstanding governors in the country and began to rise on the national stage. In an area convulsed by drug lords and armed rebels, he preached nonviolence. Inspired by the example of Gandhi, Gaviria saw the root of violence in deprivation and focused his administration on achieving greater equity and social justice (Wikipedia 2008). As part of his political campaign, Gaviria set out on a series of walks throughout the countryside to better understand the conditions and concerns of those living there. Nutrition, and more specifically, child mortality from malnutrition emerged as one of the major concerns in these communities. Although the mortality rate among children under five years of age due to malnutrition had actually declined dramatically in Antioquia during the 1980s, the rate began to rise again in the mid-1990s. The certified mortality rate had declined from 94.0 per 1,000 live births in 1980 to 17.5 in 1990, and then to 9.3 in 1994. The rate then began to rise rapidly, to 17.2 in 1999. In 2000 and 2001, the period of the gubernatorial campaign, the rates jumped again, to 32.9 in 2000 and 31.6 in 2001 (Figure 6.2). Although the reasons for this spike are unclear, violence from guerrilla and paramilitary groups continued during this period, and may be related to the increase. Table 6.1 shows the rates of chronic and acute malnutrition during this period. Relative to many developing countries, the prevalence of malnutrition and the child mortality rates were actually fairly low, but the image of a child dying from malnutrition was a powerful one. It becomes more power- ful when one can argue that such deaths are preventable. Gaviria took up child deaths from malnutrition as a signal indicator for success of social policy in his administration. Deaths from malnutrition represented a con- crete indicator in which progress could be measured, yet one for which his administration could conceivably make gains. Gaviria’s adoption of the issue raised the political visibility of malnutrition and gave his administra- tion a strong political point around which to rally social action—specifically, action on hunger and malnutrition. Other results of this prioritization were a shared focus on results among ministries and a transformation of nutrition from being simply a project to being a social issue that required a program- matic response. 104 CHAPTER 6 IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 105 Certified Estimated, including unregistered 0 20 40 60 80 100 120 140 Mortality rate (per 1,000 live births) 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 Figure 6.2 Mortality rates due to malnutrition for children under five years of age, Antioquia, 1980–2005 Source: Gobernación de Antioquia, Plan de Mejoramiento Alimentario y Nutricional de Antioquia (2006c). Table 6.1 Rates of malnutrition in children under five years of age in Antioquia, 2002 (percent) Malnutrition Region Chronic Acute Valle de Aburrá 39.7 18.0 Oriente 42.8 15.9 Suroeste 43.0 11.6 Occidente 42.2 16.5 Norte 49.0 21.2 Nordeste 47.9 20.1 Magdalena Medio 45.6 17.1 Urabá 51.4 16.1 Baio Cauca 55.0 21.4 Medellín 33.6 16.7 Overall 46.0 18.0 Source: Álvarez et al. (2004). Notes: Malnutrition is defined as below –1 standard deviation from the U.S. National Center for Health Statistics (NCHS) mean z-score. Chronic malnutrition is defined in terms of height-for-age and acute malnutrition in terms of weight-for-height. 106 CHAPTER 6 Once in office, Gaviria continued to press for action to reduce mal- nutrition in Antioquia. He turned to the minister of health to devise a plan of action. This Ministry was one of the stronger ones in the departmental government in terms of financial and human resources. Gaviria decried the fact that Antioquia did not collect indicators of malnutrition at the time and argued that “el gobierno que no mide, no sirve [the government that doesn’t measure anything, isn’t worth anything]” (DCHA). Staff initially suggested building a program to reduce malnutrition around the eight lines of action in the national food and nutrition strategy, a strategy that existed on paper but had never been implemented. He rejected that option. Influenced by experience elsewhere, Gaviria suggested replicating the Peruvian Vaso de Leche (Glass of Milk) program in every municipality. Although the implementation of a glass of milk program may have pro- moted community development and empowered community leadership, from the point of view of many technical experts, this sort of program has limited impact on the nutritional status of children in the age range of zero to two years of age, generally considered the most vulnerable period. Even if the pro- gram functions efficiently, it does not primarily target and so does not usually reach this group, nor will a “glass of milk” meet the most pressing nutritional needs of these small children. The most important factors for improving nutri- tional status of these children are exclusive breastfeeding up to six months of age and provision of appropriate weaning and complementary foods (meaning energy- and micronutrient-dense foods) beyond that age. The senior technical staff in the Ministry of Health were consequently reluctant to implement the idea. “He thought nutrition was a problem of the stomach,” they said. They urged the governor to support a more inclusive approach to nutrition. Gaviria was willing to listen to his advisers but contin- ued to press for action: “Tenemos que hacer algo [We have to do something],” he said (MM17). Initially, however, even the senior civil servants in the Ministry of Health who had been given the responsibility to “do something” were not sure what to do. Patricia Monsalve was the health services administrator. Angela Molina was an epidemiologist. Both were highly competent and experienced. They also knew how to maneuver within the government bureaucracy. Another member of the initial team, Alberto Gómez, had years of practical field experience. In interviews, all acknowledged that at the beginning, being based in the Ministry of Health, they knew little about broader conceptualizations of nutri- tion or food security. National government agencies also provided little guid- ance. The national government had various plans and programs, but there was no overarching strategy, prioritization, integration, or, indeed, political commitment for food and nutrition security. Monsalve and Molina turned ignorance to their advantage. Perhaps partly because of their own background in the sciences, they had a keen interest in learning and gathering evidence for understanding and devising practi- cal solutions to problems. They did not appear to fear their initial lack of expertise in the area and instead responded energetically to the governor’s mandate. “Initially we didn’t know anything about the topic, and we thought it was a health problem. But we concluded it shouldn’t be a health problem. [The questions became:] How do we fight hunger? How do we change minds?” (MM17).1 Instead of floundering and not doing anything, forging ahead with a health- centric program, or simply replicating what existed elsewhere, the two began a process of reflection and self-education. Their initiative was two-pronged. First, they began to find out what exactly was meant by “food and nutrition security” and to understand its determinants. Second, they began to engage others, particularly those in other ministries, to determine how others under- stood the issue and what the other ministries were doing. Monsalve and Molina uncovered helpful resources (such as the website of the Food and Agriculture Organization of the United Nations [FAO]), which provided a conceptual framework for understanding food and nutrition security, and documents from other organizations and researchers. Within this framework, they identified existing policies and programs across the departmental government that were ostensibly aimed at reducing hunger and malnutrition. Importantly, they then worked with counterparts from other sectors, such as the Ministries of Education and Agriculture, to develop a defi- nition of the problem they understood and shared. In these initial discussions, they focused on understanding and solving the problem, not on the activities of the sectors themselves. They quickly found that despite concern with high levels of malnutrition and the existence of various indicators (such as low birth weight), they did not have any information on the levels or spatial distribution of malnutrition in Antioquia. The national Demographic and Health Survey (DHS) had only recently been undertaken. Monsalve and Molina did some initial analysis of food insecurity and malnutrition in Antioquia with available data that sparked intense high-level discussions. In 2000 and 2001, they were given a separate budget, with a timeline of one year, to define and develop a plan. This was the origin of MANA. A director was appointed for MANA, bringing a more sectoral, agriculturally based perspective. Following the governor’s lead and not those of her techni- IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 107 1 The codes for the interviews are listed in the references. cal advisers in the civil service, this director believed that the “glass of milk” program should be the flagship component of the plan. The glass of milk program fit traditional concepts of a food and nutrition program, and she was able to get funding for such a program from the departmental government very quickly. Her own political interest in supporting and carrying out the gover- nor’s wishes overrode more thoughtful technical advice on how to address the problem and actually achieve the goal set out by the governor. The work by Monsalve and Molina had helped to identify the problem, but the solution chosen was at odds with what their evidence suggested. Between January and July 2002, negotiations over milk production and distribution continued with the private sector. A few events in the first half of 2002 then changed the course of MANA and gave technical advice more prominence in MANA’s development. For various reasons, including the impetus she was giving to the “glass of milk” program, which seemed at odds with a more health-centric approach to malnutrition, the director was having continuing disagreements with the minister of health. And in April 2002, the FARC kidnapped Gaviria. As part of his approach to governance, Gaviria had continued to undertake various walks, or marches, through the countryside for the sake of peace and social justice. On April 21, 2002, while he was leading a March for Nonviolence, he was kidnapped by the FARC, along with an ex-minister of defense and his peace adviser, Gilberto Echeverri. For more than a year, national and inter- national governments and groups pressured the FARC to release the hostages. In May 2003, as army troops closed in on a rebel encampment as part of a rescue mission, the FARC murdered Gaviria, Echeverri, and eight soldiers, who were also fellow hostages (PBS Online News Hour 2003). Eugenio Prieto replaced Gaviria as interim governor (April 2002 to Decem- ber 2003). MANA’s director was not so close to Prieto as she had been to Gaviria. At the same time, Prieto was prone to listen to technical arguments, and he wanted results. On both counts, the director was struggling. During the rest of 2002, the program, such as it was, continued to try to develop an operational plan. Ultimately the director was replaced. A new director, Dora Cecilia Gutiérrez Hernández, arrived in June 2003. With so much attention on the “glass of milk,” little progress had been made in defining, finalizing, or implementing other aspects of the program. In the meantime, the earlier director had managed to upset many of the partner ministries by essentially attempting to take over programs and responsibili- ties that they viewed as their own (MC27). Proceeding in this way had aroused institutional (some might say political) sensibilities over operational territory, making cooperation across agency lines even more difficult. 108 CHAPTER 6 The new director’s priority problem, then, was to win back the confi- dence of MANA’s presumptive operational partners. Gutiérrez quickly empha- sized principles of respect and transparency and of working in partnership with other agencies, not imposing outside ideas on their way of working. “In negotiations,” she said, “your word is important. We respect their ways of working, and they respect ours. And where we might work together, we ask how” (GHFV). This attitude partly reflected Gutiérrez’s own background of working in the nongovernmental organization (NGO) sector. Limited resources there required fundraising skills and the ability to collaborate with others (not simply ordering them around or defending one’s own agency’s point of view) and with communities. She excelled at forming teams and appreciating the strengths of her staff and potential partners. She also knew how to build an organization and believed that an essen- tial element of being successful was, in short, being successful—actively demonstrating to your partners and clients that you were achieving results. Taken together, these ideas transform management mentality to one that prioritizes shared results and allies, as opposed to individual ownership of accomplishments and competition with other institutions. It also elevates the role of evidence and the monitoring of management and impact to establish a shared and neutral basis for discussion of the effectiveness of collaboration. Such an approach may seem apolitical, but in fact it is not. Gutiérrez was keenly aware of the need for political support if a program was to be successful and sustained, although she marshaled that support on her own terms. “When we have intelligent politicians,” she said half jokingly, “We have to take advantage of them” (GH01). At this moment, she seemed to suggest, the political leaders in Antioquia were willing to listen to the evi- dence and work together to achieve a common goal of reducing hunger and malnutrition in the department, and they seemed less interested in promot- ing their own personal or institutional agendas around the issue. She seemed willing to press that political advantage and get them on her side, or at least on the side of the program. Although she clearly thought about the upper levels of politics, she also felt programs could lose support if they failed to reach down to the commu- nity. “The problem with Zero Hunger [in Brazil], for instance, is that it did not get down to the level of the community. MANA should be a community- level program, so that the community is aware of it, the community monitors it, and they know what they’re due” (GH01). So Gutiérrez’s approach responded to the political imperatives felt by politicians across multiple levels: MANA would not hesitate to publicize the IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 109 role of partner ministries, allowing them to share in success, and it would provide local leaders with additional resources and a visible program that reached deeply into their communities. In this way, Gutiérrez built the politi- cal support—the ownership—the program needed. Through consultation and negotiation, through securing mutual agree- ment on goals and action, she fairly well neutralized the institutional fight response that the previous director had engendered. MANA and the Ministry of Health no longer threatened the other ministries; instead they were seen as partners that could produce political benefits for the others. Specifically, even as MANA incorporated actions of other ministries into its operations and granted them the recognition they wanted for themselves or their agencies, MANA’s new approach largely allowed them to continue to operate their own programs. It thus left them control over action in their own spheres. This approach shifts the potential impact of politics, which usually means defend- ing or promoting personal or institutional goals over societal ones, from a negative to a positive force. When Gutiérrez arrived, the program was still on paper, but thanks to the continuing efforts of Monsalve and Molina, it was well designed and still engaged the interest of operational partners. Throughout this period, Monsalve and Molina had continued to bring stakeholders together, initially from the Ministries of Agriculture, Education, and Health. Significantly, these meetings were not one-offs but took place fairly frequently over a period of time. No large workshops or conferences were held, just meetings among colleagues in other ministries also working on food and nutrition issues. Col- leagues in the various partner organizations thus had multiple opportuni- ties to meet with one another privately and on a small scale to determine follow-up actions and obligations. Almost 45 formal and informal meetings were held during this initial period. In these meetings, the institutions began to discover that they were work- ing on the same issues, but their efforts were disjointed. There was no informa- tion system to discover what each one was doing. Such disorganization was partially a result of the lack of prioritization that had been given to food and nutrition. These meetings encouraged understanding and ownership. “If [you are] not present in the process of analysis, and of creating a vision, of under- standing and of capacity, it doesn’t belong to you. You aren’t committed to do anything” (MM17). In this sense, the creation of MANA was a true initiative: it was not just a statement made at a single conference or high-level meeting. Rather, the core members from the Ministries of Health, Agriculture, and Education, in particular, met multiple times to discuss issues both big and small. These 110 CHAPTER 6 meetings provided genuine opportunities for personal exchanges of ideas, perspectives, and ways to proceed. Many times the conversations were not easy, as staff from different min- istries had different ideas about what food and nutrition security meant. Through multiple individual meetings, the different organizations hammered out a shared vision, what each could do, and how they could work together. This initial group was composed of people who were already interested in and working on nutrition, with their passion perhaps making development of a sense of common purpose easier. Two of the key individuals outside the Ministry of Health, Alírio García of the Ministry of Agriculture and Marta Celis of the Ministry of Education, recalled: We had difficulties in the institutional part, with the ideas of each profession. Agriculture thought agricultural development was more important than anything else. But in food security there is also health and education. We had mistaken ideas. We didn’t yet have an idea of what “food security” was. We lacked a deeper conceptualization. We built the project together, based on . . . the mission of each [ministry]—agriculture, health, education. And so we dissolved the barriers, and problems became strengths, as we learned from one another. For this to work, you have to get people together so they live in the same house. (AGMC) Notably, this initial process of consultation and conversation occurred at the level of other technicians, not upper-level policymakers, such as minis- ters or the governor. In part this reflects the structure and capabilities of the government in Colombia and in Antioquia in particular. The civil service is composed of experienced professionals who are well respected, knowledge- able, and competent. Although policymakers may change, depending on the administration in power, the civil service carries on regardless of party. This continuity gives civil servants an understanding of policies, programs, institu- tions, and decisionmaking structures (of their own agency as well as others). In many cases, civil servants know one another, even if they work in differ- ent ministries. These characteristics—stability, competence, and institutional knowledge—mean that ministers can confidently turn to these technicians for advice. It also means that senior civil servants in each ministry are accustomed to making arguments based on technical rather than political grounds. Interestingly, outsiders might interpret disagreements among the actors as expressions of bureaucratic politics, as each organization struggled to defend IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 111 its programs and put boundaries around institutional turf. Yet in the separate key informant interviews of the four main staff members involved in these initial discussions (from the Ministries of Health, Education, and Agriculture), all said that it was the different paradigms of understanding each brought to the table that provoked the heated debates. They each were aware of maintaining the prerogatives of their home ministries, but not one mentioned politics or defense of turf as a cause of the disputes. Bringing conceptual models and evidence to the discussion helped them to stay focused on the goal they shared and were working toward and the role each ministry could play. Thus MANA’s evolution as a joint integrated program tended to mini- mize the potential for boundary wars. This process incorporated many elements known to promote successful collaboration and organizational change. Monsalve and Molina based their arguments on solid evidence. By approaching other key actors with some humility and being open to the perspectives or needs of others, they were able to encourage a common vision of the problem and potential solutions. Through the process, participants deepened their own understanding of the issue and of how their organization and programs could help while maintain- ing the integrity of their institutions. The process also resulted in genuine institutional ownership of the issue and solutions. We formalized a process of consultation, of creating a space to con- tinue the conversation, perhaps a committee, so as not to lose co- ordination of the project. Through working groups, we gained space and took action little by little, with knowledge and commitment. It was a slow process of change. Each person was protecting their own institution, but they all did work together. The process was to under- stand the issue through learning by doing. (MM17) The conversation among ministries soon expanded to include actors out- side government. Invitations to participate went to a broad range of actors who had an interest in the issue and not just those with an operational role. Monsalve and Molina included not only other departmental government agen- cies with which they might coordinate operationally but also other actors who could contribute to the program or who might have an interest in the issue. In the end, 17 governmental and nongovernmental entities formed part of a mesa de trabajo, or working group, including universities, NGOs, and the private-sector groups (such as regional development authorities or producer associations). Government organizations included the departmental Ministries of Health, Agriculture, and Education; national agencies, such as the Colombian Institute for Family Welfare (Instituto Colombiano de Bienes- 112 CHAPTER 6 tar Familiar, or ICBF); and the public university, the University of Antioquia. The group received some financial support from the Swiss government and the World Food Programme (WFP). Table 6.2 lists the partner organizations to July 2006. These other actors provided technical and financial support, insight, and expertise. At later stages these organizations and individuals often became operational partners as well. The leadership style of Monsalve and Molina, later supported by Gutiérrez, was also important: in addition to including many sectors, they appear to have genuinely listened to the views of others and ultimately devised ways to incorporate responses to the institutional needs of those potential partners. IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 113 Table 6.2 MANA’s partner organizations, 2001–July 2006 Organization name Organization type Instituto Colombiano de Bienestar Familiar (Colombian National government social welfare agency Institute for Family Welfare) Fundación Banco Arquidiocesano de Alimentos de Nonprofit religious-affiliated nongovernmental Medellín (Food Bank of the Archdiocese of Medellín) organization Ecopetrol State-owned enterprise Cornare (Corporación Autónoma Regional del Río Nare) Regional development association (Río Nare Regional Corporation) Corantioquia (Corporación Autónoma Regional de Regional development association Antioquia) (Regional Corporation of Antioquia) Comité Departamental de Cafeteros (Departmental Private-sector association Coffee Growers Committee) ReSA (Red de Seguridad Alimentaria) (Food Security National social protection program Network) Cormagdalena (Corporación Autónoma Regional del Río Regional development association Grande de la Magdalena) (Regional Corporation of the Río Grande de la Magadalena) CISP (Comité Internacional para el Desarrollo de los Italian nongovernmental organization Pueblos) (International Committee for the Development of Peoples) Ocensa (Fundación Oleoductos de Colombia) (Colombian Private-sector foundation Oil Producers Foundation) Central Mayorista (Wholesalers Union) Private-sector association Clínica Santa Ana (Santa Ana Clinic) Private health service FEDEPANELA (Federación Nacional de Productores de Private producer association Panela) (National Federation of Brown Sugar Producers) Noel Private-sector firm Nestlé Private-sector firm Fundacion Educativa El Café (El Café Educational Private-sector funded foundation Foundation) Universidad de Antioquia (University of Antioquia) Public university Universidad Pontificia Bolivariana (Pontificial Bolivarian Private university University) (continued) Interestingly, Monsalve and Molina, along with Mauricio Hernández, an adviser close to the governor, also initiated a more formal process, Situational Strategic Planning (Planificación Estratégica Situacional, or PES). PES was an approach to planning and management developed by Carlos Matus, a promi- nent Chilean economist and professor of management. Using PES, partici- pants from various institutions learned how to analyze and address a given situation. The PES approach stressed the importance of specifying initial conditions, available resources, and intended results and objectives. The course trained 114 CHAPTER 6 Table 6.2 Continued Organization name Organization type Universidad La Salle (La Salle University) Private university Tecnológico de Antioquia (Technological Institute of Public technical institute Antioquia) Universidad Catolica de Oriente (Catholic University of Private university the East) Corporación Educativa COREDI (COREDI Educational Private educational institute Corporation) Politécnico Jaime Isaza Cadavid (Jaime Isaza Cadavid Private technical institute Polytechnical Institute) Fundaunibán (Fundación Social de Uniban) (Uniban Private-sector foundation Social Foundation) Augura (Asociación de Bananeros de Colombia) (Banana Private-sector association Producers’ Association of Colombia) IKALA Private-sector consulting firm Sofasa (Sociedad de Fabricación de Automotores) Private-sector corporation (Automotive Makers Society) Programa Mundial de Alimentos (World Food Programme International multilateral organization [WFP]) Organización de Las Naciones Unidas para la Alimentación International multilateral organization y la Agricultura (Food and Agriculture Organization of the United Nations [FAO]) OPS (Organización Panamericana de la Salud) International multilateral organization (Panamerican Health Organization [PAHO]) INCAP (Instituto de Nutrición de Centro América y International multilateral organization Panamá) (Institute of Nutrition for Central America and Panama) Federación Nacional de Cafeteros (National Federation Private producer association of Coffee Growers) Colanta (Cooperativa Lechera de Antioquia) (Dairy Private-sector cooperative Cooperative of Antioquia) Land O’Lakes Private-sector firm ISAGEN State-owned enterprise Source: MM17. Note: MANA means Plan de Mejoramiento Alimentario y Nutricional de Antioquia (Food and Nutrition Improvement Plan of Antioquia). managers to analyze the strengths and weaknesses of different actors (and potential partners). It helped managers to determine what actions were nec- essary to overcome bottlenecks or take advantage of opportunities and what factors would determine the success or failure of those actions. Using PES, the participants analyzed the problem together, constructed problem trees, and determined how to respond. They also came to understand nutrition as a human right. These efforts resulted in several products, some of which were important in guiding overall actions. By April 2001, the group had developed a brief analysis that named the problem (high levels of malnutrition in children 14 years of age and younger in the Department of Antioquia) and the program (Departmental Food and Nutrition Security Plan). The analysis noted various symptoms of the problem, including high levels of poverty, infant mortal- ity, and chronic and global malnutrition; lack of detailed information on the problem and its causes; and lack of effective programs and of coordination and strategies among pertinent institutions. The analysis contained many of the initial ideas that would form the axes of action for MANA. Just creating a platform—for contemplation, planning, and action—energized the partici- pants because, despite their interest, the government had never previously talked about nutrition (MM17). The group decided to implement PES in each partner organization. The participants would identify bottlenecks on an action map and try to under- stand the cause of the bottleneck: things they could not do because they did not have money, it was beyond their authority (it required national- or municipal-level attention), or it was an issue far beyond their ability to tackle (unemployment, for example). Said Monsalve and Molina (MM17): “They built the map with their visions and their knowledge, so that they could see how their ideas were reflected. They believed in it, but everything was mixed together—[the contributions from] agricultural technicians with [those from] people from health, for instance.” These analyses often helped identify bottlenecks to action in each organization. For example, the universities discovered that their current curricula did not really prepare their students to work in nutrition at the level of programs or the community. Monsalve and Molina operated within a reasonably supportive political environment. Although Hernández was a continuous link to the governor, Monsalve and Molina also had access to the governor when they needed it. The governor was, they said, a “good ally.” He had a personal interest in the program and helped to raise funds nationally and internationally (MM17). Thus they were able to obtain the governor’s support at critical moments. That was important for gaining the support of other partners in the government, even if the other actors were not initially sure of the way forward or their IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 115 particular role, because “when the leader understands, that knowledge radi- ates to everyone” (MM17). It was also important that MANA had, even from the beginning, enough funds to get planning and then operations started. This stability created further confidence among partners from the beginning. These observations suggest that it is not critical to involve the highest level decisionmaker on a daily basis or even actually have him or her lead the initiative, but it is important that the top level take ownership and make clear that the issue is a priority for the administration. Making sure at least initial efforts have funding, even if permanent sources must be found later, can be an important demonstration of commitment. In addition, it does not hurt that others also know you have a direct line to the decisionmaker when needed! Gaviria’s management style also helped to establish networks and to get various agencies to work together. He listened to people and gathered broad support for action. Key informants suggested that despite the first director’s insistence on a milk program, Gaviria himself focused on achieving results rather than implementing any particular program, and he was open to reason. He emphasized the importance of measurement and of holding staff account- able. MANA moved forward under Prieto, who continued this orientation. At the same time, either through luck or astuteness, well-trained, well- placed, and committed civil servants were assigned to this task. In some sense, having an organizationally blank slate and not knowing much about nutrition or nutrition interventions helped. Monsalve and Molina worked to create a program that simply made sense conceptually and technically, with- out needing to defend specific programmatic or institutional prerogatives. Monsalve and Molina’s use of proven consensus-building tools was also key, as was their openness to new ideas and the contributions of others. This experience highlights the importance of personalities and leader- ship at all levels in making collaborations work. The importance of respect and collegiality that these individuals demonstrated—two human rather than technical elements—is often ignored in analyses of organizational collabora- tion. Said Monsalve and Molina, “A lot of times the plans think about the institutions and not the people” (MM17). By thinking about people and their personal and organizational perspectives, they were able to bring others together and achieve their own goals. This sort of lateral leadership appears to be critical to multisectoral col- laboration, where the leader has no direct authority over actions of other partners (Fisher and Sharp 2004). Monsalve and Molina did exhibit leadership, but it came gently (although emphatically) from the side, in ways that even Fisher and Sharp (2004) do not exactly capture (their focus still seems to be on the traditional leader of a collaborative project rather than on an initia- tive or program). 116 CHAPTER 6 IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 117 Summing Up: Getting Started Table 6.3 provides a brief timeline of this start-up phase. A close reading and interpretation of MANA’s experience with starting up the program highlights the following issues, corresponding to the main elements of the conceptual model. Internal Context • Leaders and managers appeared to share a similar style, with an approach of inclusion and co-ownership, a learning mentality, a focus on results and use of evidence, and respect for others and their institutions. • Leadership emerged at various levels and at different times to accomplish different things, from high-level authorities and political appointees (gov- ernor and director) to civil servants (in the Ministry of Health and other partner ministries). • Those responsible for designing the program set out processes to create similar understandings of the problem across sectors, a consensus on solu- tions (including how each agency could contribute), and a vision for how to get there. • The designers were senior officials with strong technical capacities and the authority to lead the design process, including the ability to call meet- ings and make initial decisions about program objectives and design. • The lead organization provided partners with incentives to collaborate— personal (own interest), organizational (mandated to make a collaboration work), and financial. External Context • The highest political authority in the department made reducing hunger and malnutrition a public priority. • Public commitment created a sense of urgency and an incentive for offi- cials across government sectors to act (rather than defending their own programs). • Higher authorities held participants accountable for achieving results. • A series of meetings and monitoring from the top maintained the sense of urgency; outcomes were defined for subsequent meetings and a timeline was set up. • Participants assigned to develop the program had a budget to undertake initial planning and design activities. • The influence of politics on the collaboration was diminished through inclusiveness, use of evidence, respect for institutional missions and pro- cedures, and incorporation of those ways into the program, instead of competing with them or trying to eliminate them. 118 CHAPTER 6 Table 6.3 Timeline for start up of MANA 2000 March Governor calls on Ministry of Health staff to develop a plan. June PES is initiated at the departmental and partner levels. November Working group’s problem analysis is consolidated. After the elections, the Health Section drafted the initial analysis, “Bottlenecks, Operations, Demand for Operation and Action: The Macroproblem of Health.” The primary problem was defined as deterioration in the health of the population of Antioquia, especially through inefficient and ineffective management of health promotion and prevention. Reducing malnutrition is first stated as a public policy goal. Analysis includes goals of reducing overall malnutrition, improving micronutrient fortification, promoting breastfeeding, and integrating care for children under five years of age (including deworming, micronutrient supplementation, good feeding practices, food fortification, complementary feeding, and nutritional rehabilitation). Implementing partners are targeted. They include the following: • Ministry of Agriculture, municipalities, and the Colombian Agricultural Institute for improving agricultural production • ICBF and NGOs for complementary feeding and nutritional rehabilitation • ICBF, food industries, and the National Institute of Food and Medicine Surveillance for increasing production of nutritious food products and improving micronutrient fortification • Secretary of Education for providing rural schools with productive, environmentally sustainable projects • Local organizations for empowering communities 2001 April Lines of action are developed through further PES analysis, with production of a situation analysis for the department. The working group designs strategies to address the problem. The Departmental Subcommittee on Food Security, a subcommittee of the Departmental Committee on Social Policy, is officially established. Requirements for participation are developed, including municipal agreements that require the formation of coordinating committees. Work to create support structures for the program is begun: this involves developing budgets, obtaining materials, creating beneficiary lists, and identifying community workers. 2002 January–July Complementary feeding component includes milk, and negotiations continue with suppliers. April–September Contracts begin to be set up for financing and operations. December First press release is made announcing the Departmental Plan for Food Security and its six programmatic components. Source: MM17. Note: ICBF means Instituto Colombiano de Bienestar Familiar (Colombian Institute for Fam- ily Welfare); MANA means Plan de Mejoramiento Alimentario y Nutricional de Antioquia (Food and Nutrition Improvement Plan of Antioquia); NGO means nongovernmental orga- nization; PES means Planificación Estratégica Situacional (Situational Strategic Planning). IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 119 • Elements of serendipity came into play: The initial director, who had a more traditional, sectorally focused management style, was replaced. The lead staff in the Ministry of Health were technically competent and inter- ested in pursuing a multisectoral approach, and they were ready with a plan when the opportunity came. Institutional Links • Organic linking mechanisms emerged from common understanding, shar- ing, and discussion of evidence. • The process of emergence took time—time to reach agreement on the problem, solutions, and vision among a core group; to reach out and gain support of others; and to deal with and respond to changes in management and the political environment. • The program moved forward with other agencies when management became more transparent and inclusive, paying attention to partner needs and operational routines. • Broadly inclusive partnering in initial stages, including in-depth discussions about vision and how each partner could contribute, strengthened the commitment by partners to the institutional links. • Systematic processes were used to gather and guide the contributions of others, helping to maintain the focus of the program as it was developed. Some other factors not readily identified in the framework established in Chapter 3 also appeared to have influenced the initiation of MANA. The strong focus on results, and not just on a routine implementation of some monitoring mechanism, stands out. The pressure from the governor’s office, as part of a campaign pledge, was constant throughout, so that staff were continuously held accountable. Both Gaviria and Prieto appear to have held ministers accountable for action. The beginnings of MANA appear to emphasize the importance of a people- oriented perspective, not just a focus on institutions. This description of start-up also suggests that Monsalve and Molina had a particular approach to getting started, and that Gutiérrez had a particular management style. These observations suggest that a strategy (a complete bundle of elements working together) was critical to success, rather than the specific presence or absence of key factors (as the description of hypotheses in Chapter 4 or summary com- ments on factors above might suggest). The approach seems to involve an inclusive process of building consensus about the problem and its solutions; of respecting the institutional mission and procedures of others and helping them see how their agency could contribute to achieving a common goal and vision; and using evidence and reason to support the discussion. These sorts of ele- 120 CHAPTER 6 ments are largely present in lateral leadership, a style that seems particularly well suited to managing complex systems (such as multisectoral programs), where the leader has little direct line authority and so must rely on persuasion and incentives to create alliances for working collaboratively. The implementation of a particular approach designed for analysis with multiple stakeholders (PES) was also uncommon. PES allowed partners to conduct a systematic problem analysis in a fairly institutionally neutral fashion to identify solutions and managerial or institutional obstacles. They con- textualized the approach by incorporating personal contributions and avail- able data. The replication of this type of analysis in other agencies led to a shared conception of the problem and appropriate action. Holding multiple small meetings with defined outcomes and follow-on steps, rather than large high-level workshops, kept momentum going. This process also allowed for a rather organic emergence of ways of working and of the programmatic struc- ture of MANA. Technical and financial capacity seems more important at this stage than operational capacity. The large role played by information has not been noted in previous analy- ses. Monsalve and Molina used the internet to find conceptual frameworks they could use. They brought in recent data on nutrition conditions to under- stand the problem and raise awareness. Their direct line to the governor may have created friction, because it allowed them to circumvent the director if necessary. And it also demonstrated the governor’s interest in hearing a technician’s point of view and getting a grasp on the extent of the problem and on goals he could use to monitor progress. The significance of senior civil servants in pushing the program forward may be telling. When a new director proved receptive to creating partner- ships with other ministries, the groundwork done by Monsalve and Molina quickly paid off. The problem analysis progressed, as did the writing of an initial strategy paper that identified policy objectives and institutional struc- ture. Thus high-level authorities may need to establish the issue’s priority and create initial political space, but mid-level technicians can then fill in this space with interventions and collaborative mechanisms to move the ini- tiative forward. Leadership, management styles, and certain approaches to collaboration thus emerge as highly influential elements here, more strongly than most of the literature on collaboration, especially in nutrition, seems to suggest. Specific personalities appear to play key roles at critical times. The lack of emphasis on these factors in the literature may be because most studies tend to focus more on the causes of failure, citing mostly institutional issues or bureaucratic infighting over turf. The MANA experience seems to suggest that inclusive leadership, expressed in a variety of ways at different times by IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 121 different individuals, may be a key factor in success that until now has been largely overlooked. The richness of these differences in leadership is notable and perhaps even critical. As the top decisionmakers, Gaviria and Prieto led by setting overall priorities and goals and by focusing others on achieving results. Gaviria’s lead- ership was important to creating the political space for nutrition. Monsalve, Molina, García, and Celis led at the technical level. They carried great weight within and among ministries as senior civil servants. They commanded respect for their technical knowledge but, within usual political limits, they also pos- sessed influence. They had the authority to call meetings and negotiate across agency lines. They were the ones who took up the challenge to fill the space Gaviria had created and that Prieto continued to want filled. And Gutiérrez, in her position as director, a political position, made many of the final operational decisions and infused MANA with an inclusive style of collaboration. She also thought and operated as a politician. Her role in managing politics was crucial to keeping the program functioning as both a program and a cross-sectoral endeavor. Gutiérrez had to manage that critical political space between the governor, ministers, and civil servants. She had to manage politics, as they might affect the program more than the policymakers (who were focused more on high-level strategy and policies) and civil servants (whose role was understood to be technical and therefore apolitical). There seems to be, then, a variety of leaderships across levels and over time. The appearance of a particular kind of leader at specific junctures in the development of a program—namely at initiation, creation, and operation —seems critical. The content and source of leadership may thus vary by the need of the program at a particular moment in time. For this reason, multi- sectoral work requires a series of leaders who are able to initiate, create (design), and operate (implement and manage) a program. Working Together: The Dynamics of Multisectoral Connections This section considers what happens after a program gets started. How do the partners actually work together on a daily basis? What makes for a success- ful collaboration? And what are the challenges? Unsurprisingly, in a complex system, it is the interaction of a variety of factors and not one single factor that seems to lead to operational success. MANA’s experience suggests key elements of such an approach, which this section describes in more detail. MANA ultimately was composed of six components, or axes of action (Fig- ure 6.3): • Community Alternatives for Complementary Feeding (Alternativas Comuni- tarias de Complementación Alimentaria). The target population was chil- dren under 14 years of age, with a special focus on pregnant women and 122 CHAPTER 6 children under six years of age.2 In addition to providing food assistance for children between six months and six years of age (in the form of a fortified snack), this component implemented a school curriculum on good dietary habits and a healthy lifestyle. It also provided dietary counseling to those responsible for school cafeterias. The Ministry of Education was largely responsible for this component. • Introduction to Health Services (Inducción a los Servicios de Salud). The main purpose of this component was to promote access to health services for children under 14 years of age and their families. Activities at the community level included promotion of breastfeeding, establishment of nutritional rehabilitation centers, integrated attention and complemen- tary feeding for pregnant women experiencing low weight gain, and devel- opment of local plans to prevent child mortality from malnutrition. The component trained staff at health facilities in food security, the treatment of malnutrition, and the Food and Nutrition Security Plan. It also helped train local leaders to assist with nutritional surveillance activities. Complementary feeding Health services Rights of the child Nutrition surveillance Agricultural production School curricula Figure 6.3 MANA’s six axes of action Source: Adapted from Gobernación de Antioquia, Plan de Mejoramiento Alimentario y Nutricional de Antioquia (n.d.). Note: MANA means Plan de Mejoramiento Alimentario y Nutricional de Antioquia (Food and Nutrition Improvement Plan of Antioquia). 2 The inclusion of children up to 14 years of age shows that MANA was not solely a nutrition pro- gram, which might choose then to focus on the most highly vulnerable age range of zero to two years of age. It was also a food security program. One of the major avenues to improving food security is education. This component’s educational curriculum targeted school-aged children to teach them about food and nutrition, especially healthy diets. • Treating Children Well (Nutrición con Buen Trato). Many parents seemed to see malnutrition or poor growth of infants or young children as a given. Until these children reached a certain age and seemed likely to survive, they did not consider these children as worthy of much attention or care. Parents paid little attention to their sick or malnourished children. This component emphasized the rights of the child and aimed to promote cultural change. MANA developed a short course that could be used to educate parents and community leaders (who would then train others) on the rights and value of children. The component also worked to strengthen the Social Policy Coun- cils at the municipality level and other community organizations. • Nutrition and Food Security Surveillance System (Sistema de Vigilancia Alimentaria y Nutricional). The information system was intended to pro- vide data for measuring the food and nutrition problem in Antioquia and for improving program management at department and municipal levels, including interaction with sentinel sites and community leaders. The data could also be used for research. • Productive Agriculture Projects (Desarrollo de Proyectos Productivos Agro- pecuarios). Although operated largely through the Ministry of Agriculture, this component also coordinated with the Ministry of Education to provide training materials for schools and the agricultural extension service. The objective of this component was to improve food and nutrition security of the most vulnerable rural families by helping them to increase and diversify household agricultural production in environmentally sustainable ways. Extension agents provided guidance about diets; food preparation, preservation, and hygiene; farm management; agricultural production techniques; and community organization. Participating families received access to credit and an initial stock of pigs and seeds to help them transi- tion to more sustainable and more productive techniques. • Educational Projects (Proyectos Pedagógicos). Working through rural edu- cational institutions and centers (including formal schools and community centers), this component designed new materials for use in classrooms and adult training. Community organization and participation were priorities. The component provided a platform for learning and improving incomes and lifestyles at the local level. The approach emphasized holistic learning and pedagogical processes. This emphasis permeated other components of MANA as well. Topics taught included business management, agricultural production, food practices, healthy lifestyles, and rights and obligations regarding health. Through community organizations, the component orga- nized and funded related Productive School Projects. Staff also advised on other educational projects that could tie in to MANA’s objectives, such as the Institutional Education Project and the Municipal Education Project. IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 123 Given the consultative process and the focus on results (and thus on making sure that all needed actions were integrated into the program), the programmatic content of MANA is not surprising. The six components (or axes) cover the main determinants of food and nutrition security at the household level. Their activities were designed to increase incomes and access to appro- priate foods, improve feeding and caring behaviors, and increase access to health services. At the same time, the program addressed some of the larger, more complex contextual determinants of food and nutrition insecurity. The program focused on education; the capacity of individuals, communities, and staff; awareness of rights, especially of the child; and the creation of a management information system. In the spirit of collaboration and integration (which one might also inter- pret as a political calculation to minimize interministerial conflict), instead of trying to re-create what existed elsewhere, staff built on and integrated exist- ing programs from partner ministries. This was particularly true of the agricul- tural, pedagogical, and complementary feeding components. The approach was likely due to various factors, such as the development across a diverse group of partners of a holistic vision of food and nutrition security, which included the importance of education and agricultural production and the inclusion of strong proponents of agriculture and education in the core team who argued for their programs’ inclusion. Another significant factor was the apparent understanding by the Ministry of Health staff that their own institutional advantage lay in health-related actions and that the design and management of other components was best left to others. Through consultation, staff from the various ministries thought about how each could contribute. Ulti- mately MANA emerged as a coordinating body or umbrella organization that integrated these different components and oversaw implementation at the municipal level. MANA had three organizational levels: the overall coordinator, the head- quarters coordinators of each program component, and the regional teams (composed of the contracted operational partners, who were responsible for delivering and implementing the program components in the municipalities). Each partner on the regional team was responsible for one of the six compo- nents, or axes. The operating partners affiliated with each component (those implementing the education axis, for example) also met from time to time. In addition, a separate team was responsible for working with each municipality to build community capacity to demand and implement the program. A close study of MANA sheds light on the relative importance of the various internal and external factors suggested by the literature and also uncovers other factors. In MANA, as in Programme de Renforcement Nutritionnel (Nutrition Enhancement Program, or NEP) in Senegal (see Chapter 5), collaborative and 124 CHAPTER 6 visionary leadership along with structural flexibility and the appreciation of the efforts of partners was key. Also, as in Senegal, involvement of local communities seemed to be an important element of success. Previous analy- ses have tended to overlook these factors in explaining success in working multisectorally. This section describes other factors that emerged as being important to operational success. Leadership from Below? Involving Civil Servants The participation and leadership of technical civil servants contributed to MANA’s initial political and organizational success. As explained in the previ- ous section, preliminary discussions about program content occurred largely at the technical, rather than the political, level. Thus the principal issues of coordination and technical disagreements were largely resolved by the time those questions reached ministers. The cadre of civil servants also provided some stability. Even if ministers changed, competent technical counsel and institutional memory remained. Even more important, the governor him- self had set the policy goal—to reduce malnutrition—so that even if there were changes of ministers, the new ministers still would have to respond to him and work to achieve the continuing, overarching policy objective. This accountability provided an incentive for ministers to achieve results. At the same time, MANA provided the administration and ministers with a signature program through which they could increase visibility in communities and claim success without much additional administrative effort. This approach differs from an organizational structure in which a higher political authority, such as the prime minister, might guide the process or manage an institution or coordinating body. Instead, coordination largely took place at a technical level. No additional high-level coordination seemed necessary, because the ministers knew that they would be held accountable for results. So rather than having the prime minister lead the program, it can be sufficient to know that nutrition is on the prime minister’s (or governor’s) agenda and ministers will be held accountable for results. Creating Vision and Building Commitment through Communications Strategic communications raised awareness of the problem among a wide range of stakeholders and built widespread commitment. When MANA began, government, families, communities, and professionals did not consider food or nutrition security a priority. A national food security strategy existed but was dormant. As Gaviria raised the profile of hunger and malnutrition and MANA began to come together, many people—mayors, for example, and even doctors—did not believe that malnutrition was a priority problem. One of MANA’s first steps IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 125 was thus to gather data on the prevalence of malnutrition and malnutrition- related deaths. The Ministry of Health used a more inclusive cutoff to deter- mine prevalence of malnutrition than that most often used internationally (children whose Z-scores were more than 1 standard deviation below the mean National Center for Health Statistics [NCHS] standard rather than 2 standard deviations below). This change had the effect of including not only those who were traditionally considered malnourished but also those at risk. Including the vulnerable is certainly technically defensible from a public health point of view, and it also served to highlight the extent of the problem and encourage urgent action. MANA communications publicized the program through a variety of media, and staff made presentations wherever they could. Staff discussed their program and presented statistics on malnutrition in Antioquia to politicians, bureaucrats, and civil society organizations in conventions, professional meetings, and regional assemblies. They also contacted the mass media, the private sector, and universities. As a result, after only a few years of operation, community residents, students, municipal leaders, government ministers, parliamentarians, and other partners (such as those from universi- ties or development associations) knew about MANA, its underlying analysis, and action plan. The communication strategy had a few key elements. First, MANA tar- geted those directly concerned with interventions to combat hunger and malnutrition as well as those who could indirectly influence the success of those interventions, such as regional development associations. The idea was to get political society at large to view nutrition as a development priority. Second, the logo designs were lively, easily accessible, and memorable. The MANA logo itself was a stylized family standing superimposed on an out- line of a map of Antioquia (Figure 6.4). Third, MANA put the logos on almost all program materials, including vehicles, food, and publications. MANA became well known throughout the department, especially at the community level. Fourth, MANA communications supported each component individually across ministries. For example, MANA produced pedagogical materials, including cur- riculum guides and workbooks, for the educational component. Staff prepared and shared documents among partners, thereby encouraging a team mental- ity. And fifth, the design guidelines allowed the seal of the Government of Antioquia and the Health Directorate to appear, but not that of any particular ministry. At the same time documents could include the logo of the relevant operational partner. This practice reduced intragovernmental conflict over credit for the program and increased buy-in from nongovernmental partners. MANA also produced compact discs with PowerPoint presentations to describe the program, its progress, and its achievements. Staff could use these 126 CHAPTER 6 presentations to explain the program to potential partners, municipalities, and community residents or for reporting to ministers and parliamentarians. This communications strategy greatly increased the visibility of MANA throughout the department. MANA became known not only for its logo but also for its actions and its presence in the community. The materials that MANA produced also became integrated into the program operations in other ministries. Experience with Collaboration MANA was new but was able to build on existing programs. Some partners already had a history of collaboration. Importantly, since 1996, the Ministry of Agriculture and the Ministry of Education had worked together on school agriculture projects. As a result, they had already worked through many of the operational challenges posed by cross-sectoral collaborations. Specifically, they knew about the importance of creating a shared vision and ownership among partners rather than focusing only on the technical aspects of a project: [In our initial program] we had a problem with school gardens. We gave inputs [to the schools], but they were wasted. They planted the gardens with corn or beans, but then they abandoned them. The people didn’t take ownership. And the gardens died when the school let out because [they thought] it belonged to [the agricultural exten- IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 127 Objectives: • Ensure food and nutrition security • Prevent deaths from malnutrition Concept developed: 2000 to 2001 Piloted: 2004 Expanded: 2006 Figure 6.4 MANA: Program development and general logo Source: Gobernación de Antioquia, Plan de Mejoramiento Alimentario y Nutricional de Antioquia (n.d.). Note: MANA means Plan de Mejoramiento Alimentario y Nutricional de Antioquia (Food and Nutrition Improvement Plan of Antioquia). sion service] or the Ministry. And so the agriculture sector saw they needed to involve Education, because it wasn’t Agriculture’s problem but one of community education. Then we co-financed a project. Agriculture put in money, and Education put in money. And together we trained teachers and agricultural professionals. The agricultural technicians learned to be teachers. And the teachers learned about agriculture. (AGMC) The first years of this joint program, however, depended almost entirely on the vision and goodwill of two staff members, Alírio García in Agriculture and Marta Celis in Education. For them, MANA now presented an opportunity to get funding for their program and so institutionalize it: To accomplish our vision, we had to find the resources somewhere. And there was money in Health. Well, we said, let’s make an agree- ment. And so that’s where we began. [The Ministry of Health] invited us because they had a food security program. And they came to Agriculture, because Agriculture had something to do with produc- tion. And Agriculture included the agriculture-education project. We started up a unit to work together. (AGMC) In perhaps an amusing extreme of participatory ownership, they named the new initiative the “healthy productive education project,” basically using the title to cover the bases with all three ministerial “owners” of the program. Organizational Placement of MANA In the beginning it was uncertain where to house the coordinating unit for MANA, as it included programs and staff from the Ministries of Health, Education, and Agriculture. Ultimately, for reasons of available space and administrative sup- port, MANA was placed in the Ministry of Health. Nevertheless, although the coordinator of MANA was technically under the minister of health, she actually reported directly to the governor. In fact, the director was the only official staff member of MANA. All other staff were on loan from participating institu- tions. Interestingly, this arrangement contributed both to independence and interdependence. Staff technically still responded to their home institutions, but for the system to work, they had to work together. Having the leadership staff work out of the same program office made this task easier. Building Multisectoral Collaboration: Weaving the Pieces Together Although the vision was developed under leadership from the Ministry of Health, the structure that emerged built on and integrated activities from 128 CHAPTER 6 other sectors. Continuous discussion kept all partners on the same page in terms of understanding and action. Participants came to recognize the intrin- sic value of each existing program and saw how to integrate their work into MANA’s axes of operation. As part of their analysis, partners identified other components that needed to be established or strengthened. This experience adds an important nuance to the statement that the capacity to carry out obligations is essential to multisectoral collaboration. MANA shows that the type of capacities needed will vary. Participating agen- cies play different roles and so have distinct capacity requirements. The important thing is that the role be appropriate and the agency understand what it is to do and be capable of carrying out its responsibilities. Organiza- tions do not have to have the same capacities or undertake all roles. But when an ability to fulfill an obligation is weak, the collaboration must find out how to strengthen the participant or work around the problem. In the case of MANA, in the end no agency had to stop what it was already doing. In fact, participants saw how working with MANA could improve the efficiency of their own programs to more effectively reduce hunger and mal- nutrition. They would then carry that conviction back to their home institu- tions. In fact, this synergy was one of the selling points of MANA to ministers outside the home base of the Ministry of Health: working through the MANA structure allowed them to achieve their own program goals at lower cost by integrating their staff into MANA or by coordinating with MANA more closely at the local level. Agricultural extension agents, for example, operated the Productive Agriculture Projects to improve sustainable agriculture and house- hold food security and nutrition under MANA’s umbrella, which obtained complementary support from the School Nutrition component. This focus on results helped shape and, in some sense, limit the conversa- tions. Instead of arguing over the programs themselves, the focus on results kept attention on three main questions: (1) How do we reduce malnutrition in Antioquia? (2) How does my program contribute to that? (3) How can it be modified to improve its ability to help achieve that result? This focus on results then more easily led to conversations across sectors about how they should work together to achieve those results. Of course, such programmatic articulation is not easy. As the history of multisectoral work in nutrition shows, what makes sense conceptually is difficult to put into practice operationally. This holds true not only at the ministerial level, where most analyses focus, but also at the local level. Community residents and local partners noted this problem during a feed- back exercise. They noted that, despite having mechanisms planned out on paper, in practice the different implementing partners still lacked sufficient articulation and coordination. Partners scheduled training sessions of differ- IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 129 ent components at the same time, for instance. Municipalities did not always fulfill their obligations to the program. Staff would not be available to work with partners, or they would not provide suitable space (such as kitchens) for program operations. Despite communications efforts, community acceptance of the program could still be weak. Said one participant: “We doubt whether food security is a relevant priority for the school. Boys and girls don’t eat salad!” Another, speaking about the educational component, complained: “We get mixed up with the Productive Project. We don’t have time to take care of chickens! You really have to get rid of the chickens!” (Gobernación de Antioquia, Plan de Mejoramiento Alimentario y Nutricional de Antioquia 2006b). Sometimes the issue was operational, a constraint that is just as important to the success of a program as conceptual design or institutional linkages. In Antioquia, for instance, continuing violence in the countryside made program operation difficult. And many housewives, the main target of some of the program components, could hardly find time to travel the long distances for training. Building up Carefully By building on existing programs, MANA avoided the wholesale start-up and imposition of a new program, which might have provoked significant resis- tance from partner organizations. According to Monsalve and Molina (MM17), this approach had a significant political advantage: they could build on what already existed and take advantage of current capacities and legal frame- works. It also followed a principle of lateral leadership—of valuing the con- tributions of others and of being humble with regard to what we ourselves know. At the outset, however, some programs were more developed than others. Complementary feeding and the agricultural production and educational projects were already under way, but the introduction to health services and the nutritional surveillance system were weak. And the component having to do with the rights of the child did not exist at all. MANA leadership, however, made it clear that these last two components were important, and the promotional materials treated each component almost equally. Nevertheless, MANA did not try to get all elements in place and at the same level of operation before introducing the program into the municipalities. MANA was allowed to grow somewhat slowly and to implement components at different rhythms, in accord with resource availability. This approach was partly a result of constraints on human and financial resources, but it nevertheless seems a wise way to deal with the problem of uneven capacity and readiness. The approach allowed MANA to learn and adapt over 130 CHAPTER 6 time and so enhanced its probability of success. The team structure promoted learning, but the team could also decide how and when to implement certain components in specific communities, depending on local capacity and needs and the strengths, capacities, and opportunities of the operating partners. Paying Attention to Partners: Incentives for Collaboration An obvious observation, but not one always taken into account, is that effec- tive multisectoral collaboration requires mechanisms to enable it. MANA was innovative in the design of its mechanisms and was careful to establish and maintain the institutional and personal links needed to make them work. MANA’s coordinator, Dora Cecilia Gutiérrez, cared deeply about creating allies. She recognized that personal relationships and incentives were impor- tant to creating sustainable links. She considered how to increase benefits to partners while decreasing their costs, a point noted in Chapter 3. She thus tried to create reasons for agencies and partners outside the Ministry of Health to work with MANA. In essence, Gutiérrez constantly asked three strategic questions: (1) What do I need to be done to achieve organizational goals? (2) Who can do it? (3) What can I offer them in exchange for their collaboration? In interviews, Gutiérrez stressed that MANA “is everyone’s program.” She considered how partners could advance MANA’s interests, and in fact turned down or ended some collaborations where they were not useful. From her perspective, the organizational success of MANA rested on getting ownership and alignment among the partners and having them cooperate in the imple- mentation of the program. “The partners are my children,” she said. “I have to take care of them” (GH01). But how could MANA, located in the Ministry of Health, get other minis- tries to work in a collaborative structure? One answer was for MANA to man- age and implement the programs of client ministries. MANA would function as a service provider. Gutiérrez sat with each minister, explaining to them how working through MANA could help the ministry and the minister achieve institutional objectives. MANA would carry out a ministry’s program, using the available funding, but merge it in a coordinated fashion with the other program components from other ministries. By letting MANA deliver their programs, ministries would profit from synergies both in terms of increased impact of their programs and, through integrative efficiencies, lower delivery costs. MANA would handle program management but still report results and financial accounts to the base ministry. The base ministry would thus continue to get credit and also fulfill its legal, institutional, and political obligations. This last point reduced the potential for conflict over institutional prerogatives and jealousies. IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 131 Monsalve and Molina described it this way: “Then the other programs keep going well, because they need MANA. And each institution gets its recogni- tion, according to its institutional interests. A main fear was that we weren’t going to give their institution the recognition it deserved. But the other organizations added their own goals, and that increased confidence and also helped to focus on results” (MM17). In addition to working directly with other government programs, MANA contracted with universities, development associations, and consulting firms to implement program components. Certainly part of the reason organiza- tions wanted to collaborate with MANA was funding. But MANA also allowed each implementing organization to adapt the activities under contract to be consistent with its own goals. Outside of some core principles and activi- ties needed to meet program objectives, MANA and the operating partner adapted the contract to the objectives of the partner and to local needs and strengths. Such flexibility showed institutional respect and also encouraged risk taking and learning. In one such example, MANA coordinated with the Food Security Network (Red de Seguridad Alimentaria, or ReSA). This national program aimed at providing technical assistance, including education, and some inputs to rural and urban families, so they could produce some of their own food. The target populations of MANA and ReSA overlapped. By working together the two pro- grams could avoid competition, reduce costs, and expand their reach. While the ReSA budget financed some of MANA’s activities, ReSA benefited from MANA’s infrastructure in delivering assistance and inputs. [What we do at ReSA is] within the bounds of what we want. Not to create more lines of action, but to join forces from the social point of view, not duplicate work. MANA and ReSA give parameters for the work. But it is a process that we are building, taking advantage of the various institutions. No one is the absolute owner. And each operating partner is different. And that is the advantage. They have allowed for a lot of participation. So we are structured to allow for that, and we don’t tell people “how” to do something and make them do it, because we don’t know how. We know “what” to do, but we don’t know “how.” (PPFG) MANA also worked with ICBF, the national program composed of 34 different social protection programs. ICBF had two programs similar to those of MANA in terms of content and target groups: complementary feeding and nutritional rehabilitation, the latter of which is one of the activities under MANA’s Intro- duction to Health Services component. The staff of ICBF realized that they 132 CHAPTER 6 could join forces with MANA and save money, using the savings to expand or intensify their program for beneficiary children. As a result of working with MANA, ICBF was able to more than double its coverage. This occurred partly because the two programs could compare beneficiary lists and uncover dupli- cate beneficiaries and “ghost children” counted by the municipalities (which managed funds at the local level) (ICBF21). In another example, MANA looked to the universities for in-depth tech- nical knowledge of food and nutrition security. By working with MANA, the nutrition department at the University of Antioquia gave faculty and students important field experiences. Feedback from those experiences then led to revisions in the curriculum of the school. MANA’s target beneficiaries of poor smallholders also overlapped with those of the coffee growers’ association and the regional development corporation, yet the coffee growers’ extension activities had greater reach and was more effective than the government service. By working together, the three organizations could reach more of the target population more effectively and at lower cost than by working separately. Fiscal Space MANA also benefited from disorder at the national level. The national gov- ernment had designated a special tax to finance activities in health at the departmental level, but did not have a clear plan on how to spend it. MANA was able to make the case that investing some of these funds in MANA would be wise. The composition of financing, though, shifted over time. In 2002, MANA had funds from Ecopetrol, the state-owned oil company; royalties from petro- leum production; and funds from the treasury designated for public health, along with an in-kind contribution from the Ministry of Agriculture. In 2003, the primary funding sources were oil revenues and public health funds. By 2004, however, MANA was beginning to gain its own budget lines, in col- laboration with the ministries—but it took years for MANA to gain this fiscal space. Creating Capacity: Supply and Demand Those who put MANA together realized that for long-term success, capacity and competence had to exist at all levels of the program. One of the opera- tional partners, for instance, divided this conceptualization into three levels, with administrative, operational, and political components: “The agreement between MANA and the operator is the first level. The decisions on the appli- cation of the methodology in collaboration with the community—the opera- tional part—that is the second level. And the third level, to work with the IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 133 community to strengthen the political actors, that is not administrative or operational but political” (BTFG). Sufficient capacity needed to exist on both the supply and the demand sides, that is, both among the organizations supplying the program (such as the operating partners) and those demanding it (such as the municipalities). Involving the communities was seen not just as important to implementation but had a political rationale as well, to provide some benefit to local leaders and gain support for the program. As noted above, MANA’s organizational structure had three levels: the overall coordinator, the headquarters’ coordinators of each of the program components, and the regional implementing teams. These teams, composed of different partners responsible for implementation of different program components, worked together across a cluster of municipalities. The demand side mainly consisted of the municipal mayors and their staffs, who coordi- nated the components at the local level. But MANA not only supplied the program to municipalities, it also aimed to support the demand side in order to strengthen the overall program. A separate team, also part of MANA, worked with each municipality to build local imple- mentation capacity. They worked to strengthen community participation, design local food and nutrition plans, and assist mayors and local staffs to coordinate the components of MANA at the municipal level. This support was important because few municipalities had the knowledge or skills to put together such a plan. By accompanying municipalities in their dealings with MANA, they lessened the risk of failure. MANA and communities were genuine partners. MANA’s approach to linking supply and demand was one of incentives rather than imposition. MANA staff visited each municipality to tell them about the program and offer them a chance to participate in it. The program did not require them to participate, however, in contrast to the usual way a centralized program rolls out or scales up. If the municipality chose to participate, MANA then required each municipality to comply with municipal laws, identify local counterpart staff, and come up with a municipal food and nutrition security plan. This approach encouraged, indeed required, local ownership. But the decision was the municipality’s, not MANA’s. Although the financing for program components came from MANA, local leaders had to put forth at least a minimum effort to understand the program and integrate it into social policies and programs at the local level. Each municipality had to have a functioning Social Policy Committee (Comité Municipal de Políticas Sociales, or COMPOS). COMPOS had to be widely representative of the community and to include civil leaders, unions, police, religious organizations, and the private sector. A coordinating com- mittee of COMPOS consisted of various government ministries (Agriculture, 134 CHAPTER 6 Health, Education, and Environment) and MANA. COMPOS also had a supervi- sory committee to oversee and assess the activities of MANA and the coordi- nating committee. Some municipal leaders still turned down the opportunity to participate in the program. This refusal was likely because of insufficient understanding of the program at first and because participation required time and some local resources. Some municipal staff complained that MANA took up too much staff time. For instance, the municipality usually had only one individual to serve as the contact with all the institutions and MANA staff and to manage delivery of inputs. But soon those municipalities who had chosen not to participate saw MANA activities going on throughout the department, as did their residents. In a few years MANA had expanded to include all municipalities in Antioquia. Community-Driven Multisectorality? Involving and Supporting the Community Staff and operating partners described MANA’s strategy with remarkable con- sistency, suggesting significant, and successful, efforts at developing a com- mon understanding of the program. The program implicitly encouraged insti- tutional articulation at the community level through a continuous process of reflective activities (such as workshops and training courses) and fieldwork. Cutting across components and regions, operational partners described a similar sequence of activities and a consistent emphasis on community trans- formation, including community-level empowerment, ownership, and action. The actual process to develop this common understanding across levels took years and included the following stages: • Year 1. Raising awareness; forming groups for action, advocacy, and imple- mentation; and training trainers (such as local leaders and teachers) • Year 2. Strengthening the activities of year 1 through continuous action • Year 3. Improving organization and learning and disseminating lessons The previous section described how MANA teams worked to create a sense of ownership from the beginning and to build capacity and empower municipal leaders and organizations. Other activities further supported multisectoral action. The local school, in particular, seems to have emerged as the articu- lating platform for activities, and the influence of educational approaches is evident in MANA’s emphasis on learning and capacity building. Social/Community Mobilization: Establishing Understanding, Urgency, and Community To establish understanding, operational partners started by inviting everyone in the community to an informational meeting. This meeting served to intro- IMPROVING FOOD AND NUTRITION SECURITY IN ANTIOQUIA 135 duce the community to the program and to answer any questions. Once the municipality and MANA had reached agreement, operational partners held additional meetings with individuals and community organizations to provide further detail on specific components. MANA staff and operational partners emphasized the importance of com- munity capacity and involvement as critical to successful implementation. The frustration of the operational partners in municipalities where communi- ties did not fully understand the program and their roles in it or where they did not have the capacity to implement it was palpable: Sometimes the community doesn’t understand. They say, “You are from Medellín, so we have to do something for you.” And so they gather a bunch of people, but they aren’t the right people. They don’t under- stand why we’re there. And in some communities the officials aren’t from there. There’s no continuity. MANA’s operational structure is a lot bigger than the operational capacity of the municipality. It is a strength that becomes a problem. Everything is very complex [in MANA, but then it lands in] a structure that is as simple as the municipality’s. We have learned it is a prob- lem, but we haven’t learned how to solve it. Although some municipalities are organized, others are not. They don’t have very solid internal structures. The person at the level of the community has to have a lot of experi- ence. They are the ones that make the work happen at the local level. At the rural level, the teacher has a lot of influence. [The teacher is] a key person. (BTFG) MANA also worked in areas that had been ravaged by violence, especially in battles with the FARC. Part of the government’s strategy was to invest in these areas and so address the issues of poverty and social inequity that had historically driven residents to support the FARC. But, according to the focus group members, the presence of so many NGOs and government programs meant that many communities were rather indifferent to MANA. They had no sense of urgency or interest in engaging a new program. In some of the municipalities where armed conflict has been the great- est, there are displacements, which have induced passivity, since many organizations now work there. (OrienteFG) And so they receive and receive, but they don’t even know what they’re receiving. So some sectors are not committed. They are saturated with what they have and aren’t interested in the project. (BTFG) 136 CHAPTER 6 Many times a whole lot of people arrive to help, but the