Cuna Más: Peru's Early Childhood Program PDF

HKS Case 2115.0 January 29, 2018 Cuna Más: Peru’s Early Childhood Program Struggles to Maintain Quality as It Scales Up Introduction On a spring morning...

HKS Case 2115.0 January 29, 2018 Cuna Más: Peru’s Early Childhood Program Struggles to Maintain Quality as It Scales Up Introduction On a spring morning in 2015, Lucia, a community worker in the Peruvian rainforest, set out on her weekly trip to meet two-year-old Hector and his parents. The walk to Hector’s house took Lucia several hours along a rugged dirt path that trailed a tributary of the Amazon. When Lucia arrived at Hector’s one-room dwelling, set in a small clearing in the dense jungle, she was greeted by a beaming child and his parents. All three welcomed her with open arms. As in past weeks, Lucia had come with a backpack full of children’s books, puzzles and games. She spent an hour with the family, playing with Hector, read- ing to him, and showing his parents the benefits of stimulating children with play. Soon, the conversa- tion turned to the importance of immunizations and of maintaining good hygiene. At the end of the visit, Lucia bid goodbye to Hector’s family, left a toy behind, and promised to return next week. Lucia was one of thousands of home visitors as part of Peru’s ambitious national program on early childhood development, Cuna Más. President Ollanta Humala, who had run for office on a platform of social inclusion, launched Cuna Más (translated: more than a cradle) amid much fanfare in 2012. The program was designed to improve the cognitive and emotional development of children below the age of three, living in poverty. Cuna Más provided two services: weekly home visits in rural areas and free, a quality childcare in urban areas. The government’s aim was to scale nationwide. Despite the logistical challenges of reaching rural communities in a geographically diverse country—from communities resid- ing high up in the Andes to those deep in the Amazon jungle—Cuna Más grew at a rapid clip, increasing coverage by 50 percent or more each year between 2012 and 2015.1 Yet, in early 2016, Jorge Luis Fernández, the executive director of Cuna Más encountered a formida- ble set of challenges in growing the home visiting program. Amid political pressure to increase coverage quickly, he had to ensure that the program continued to deliver a quality service—even as Cuna Más a For the purpose of this case the term Cuna Màs refers to just the home visiting element of the program. This case was written by Anjani Datla, Senior Case Writer in collaboration with Julie Boatright Wilson, Harry S. Kahn Senior Lec- turer in Social Policy at the John F. Kennedy School of Government (HKS), Harvard University. We are grateful for translation support provided by Patricia Garcia-Rios, Multimedia Producer at HKS. Funding for this case was provided by the Bernard van Leer Foundation. HKS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorse- ments, sources of primary data, or illustrations of effective or ineffective management. KS1251 Copyright © 2018 President and Fellows of Harvard College. No part of this publication may be reproduced, revised, translated, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means without the express written consent of the Case Program. For orders and copyright permission information, please visit our website at case.hks.harvard.edu or send a written request to Case Program, John F. Kennedy School of Government, Harvard University, 79 John F. Kennedy Street, Cambridge, MA 02138. staff navigated the delicate task of convincing communities to adopt a program that, at its heart, relied on the willingness of families to let strangers into their homes. Fernández was nonetheless hopeful. In many instances, Cuna Más was the first social program to reach remote, vulnerable populations, such as Hector’s Amazon region. But some civil society groups that had initially hailed President Humala for focusing on early child- hood now argued that the program was hobbled by operational challenges. They said two factors threatened the long-term sustainability of the program. First, a chronic lack of human capacity in rural areas was affecting Cuna Más’ ability to deliver home visits effectively. Second, because the central gov- ernment managed the program from Lima, it was neglecting to build the capacity of local governments, which were in a better position to understand the needs of local communities. As the April 2016 elections loomed on the horizon a crucial question about Humala’s flagship pro- gram lingered. Could the promising program deliver with quality at scale? The Tale of Two Perus Starting in the mid-1990s, Peru, which had recently emerged from a decade-long violent insurgency, experienced extraordinary economic growth. Fueled by lucrative investments in extractive industries, cotton and tourism, the country’s gross domestic product grew on average 6.5 percent every year from 2003 to 2012, more than most other emerging economies in the world.2 Although the overall poverty rate declined dramatically—from 58 percent in 2001 to 26 percent in 2012—Peru suffered from large economic and ethnic divides (see Exhibit B).3 There were stark differ- ences between the bustling capital of Lima and the rural, impoverished areas of the Andes highlands and Amazon forests. The World Bank estimated that a resident of Lima earned 21 times as much as a resident of Peru’s rural areas, populated by the country’s many indigenous groups, where the poverty rate was as high as 54 percent.4 High levels of poverty were particularly problematic for children. A third of all rural Peruvian chil- dren suffered chronic malnutrition (see Exhibit C for more early childhood statistics). With poor schools and a dearth of adequate health services, many civil society groups in the country were concerned that despite vast economic gains, current generations of low-income Peruvian children, who remained mar- ginalized and geographically isolated, would not be able to rise out of poverty. Although the country had made significant headway in reducing infant mortality, Peru still had to address health and education gaps among children. When President Ollanta Humala rode into power in 2011, he was among a new breed of Latin American leaders who recognized that inequality began early. He favored a strategy based on the premise that sustainable economic growth had to be inclusive. Fol- lowing the lead of governments like Brazil and Chile, Humala spearheaded the creation of a bold new program: Cuna Más that would nurture the cognitive, emotional and motor development of very young children. Cuna Más 2 of 18 HKS Case 2115.0 Cuna Más: An Evidence-Based Model The Cuna Más home visiting service arrived in Peru by way of a British doctor working in Jamaica in the 1980s. Sally Grantham-McGregor noticed that while both the low-income and middle-class infants she treated grew at roughly the same pace in their first months, once those same children reached tod- dlerhood, there was a marked difference in physical and mental growth between the two groups. “When I went to their homes, I saw that [children from low-income families] weren’t getting the right stimulation,” said Grantham-McGregor. “The children were doing nothing. They had no toys or books. The whole family lived in one room…. I saw that mothers, depending on their own socio- economic background, interacted with their children very differently when it came to playing a sorting game. The educated middle class women were naming concepts as they played, explaining to their chil- dren what goes where and why. The poor women just watched their children try to sort the toys and just kept saying ‘no, no, no’ when the child sorted incorrectly. The difference was phenomenal.”5 Grantham-McGregor wondered if there was a way to adapt elements of two early childhood pro- grams that were gaining wider acceptance in the United States: the Nurse Family Partnership and Head Start. The Nurse Family Partnership in operation in some parts of the US since the 1970s sent nurses on home visits to poor first-time mothers. And the Head Start program that began in 1965 offered compre- hensive childcare and early education services for poor families at centers staffed by trained profession- als. Grantham-McGregor knew that a center-based model was not feasible in Jamaica and decided to test home visits conducted by health workers. In 1986, with the help of colleagues at West Indies Uni- versity, Grantham-McGregor conducted a small experiment. The researchers identified roughly 130 fam- ilies in one of the poorest parts of Jamaica’s capital, Kingston, and randomly divided them into four groups. For one group, health volunteers visited the homes of mothers with toddlers and taught them how to stimulate their children by playing with them, talking to them and praising them. The second group received the same intervention as well as nutritional supplements. Both these groups were com- pared to the remaining two groups, one of which received just nutritional supplements and another that received nothing—neither the home visits nor the nutritional supplements. The program ran for two years. More than two decades later, when the researchers followed up on the toddlers who were now grown adults, they found that those who received home visits, with or with- out nutritional supplements, were more educated, less prone to violence, and earning 25 percent more than those who received just nutritional supplements or no intervention at all. In fact, the higher in- comes placed adults in the treatment groups on the same level as their peers from better off families, demonstrating unequivocally that investing in children when they were very young yielded long-term economic gains for the country. In development circles, the Jamaica study was a rare occurrence: a simple model, tested rigorously, with proven long-term benefits. The appeal of adapting the model to other countries was immediately evident. Countries across the world including Colombia, Peru, Bangladesh and India started programs on early childhood that were based on Grantham-McGregor’s work. As impact evaluations in the different countries continued to show the value of such programs, questions of how they could be scaled up dom- Cuna Más 3 of 18 HKS Case 2115.0 inated the discourse. In Colombia, for example, researchers leveraged existing administrative capacity of the nation’s conditional cash transfer system to conduct a pilot home visiting program (see Appendix 1 for a comparison between Colombia and Peru’s early childhood programs). Early results of the program were revealing. A randomized controlled trial showed that families that received home visits changed behavior, investing more time and resources in stimulating children at home. With Cuna Más, Humala prioritized the idea of national scale up by making early childhood devel- opment an integral part of the government agenda on inclusive growth. Peru would create a new na- tional program dedicated to early childhood development. It would commit significant resources, and employ a results-oriented framework to ensure that Cuna Más delivered results. That a national leader was willing to stake a claim on this issue drew the attention of international agencies, foundations and Peruvian civil society. A wide variety of organizations pledged their support to Cuna Más by contributing funding, technical support and capacity-building services. The Inter American Development Bank, UNICEF, the Open Society Foundation, and the Bernard van Leer Foundation were among the first international donors to provide financial support. Religious and community groups across the country announced their enthusiasm for the venture. Salgalu, a social responsibility organiza- tion, and Red Innova, a group of former civil servants working to build the capacity of local governments, all supported the initiative. Social Inclusion in Motion In October 2011 Humala created the new Ministry of Inclusion and Social Development (MIDIS) to bridge the health and education attainment gaps associated with poverty and marginalization. The new Ministry’s mandate was to coordinate the government’s efforts in the highest poverty areas. It would direct five social programs, some of which were already in operation, and others such as Cuna Más and Pensión65, would either be revamped or developed from scratch. b 1. Juntos, a conditional cash transfer program ; 2. FONCODES, a social investment fund; 3. PRONAA, the government entity for distributing food to vulnerable groups (later called Qali Warma); 4. Cuna Más; and 5. Pensión65, pensions for persons over the age of 65 living in extreme poverty. MIDIS was a vital tool in Humala’s “Make Growth Inclusive” strategy and would be important in co- ordinating the five social programs, as well as integrating them with activities of related programs (for c example on primary education and nutrition) undertaken by other government agencies. Humala wa- b Juntos was started in 2005 and delivered to mothers living in extreme poverty under the condition that they send their chil- dren to school and take them for regular health check-ups. c MIDIS has two arms. The first is programmatic as described above and the second is focused on planning, multisectoral coor- dination and evaluations. The second part of MIDIS wrote the Make Growth Inclusive strategy and pushed for program moni- Cuna Más 4 of 18 HKS Case 2115.0 gered that it was this integrated and coordinated approach that would finally break persistently high poverty in Peru. It was novel for a government ministry to create an early childhood program as one of its core ser- vices, rather than mandating collaboration between traditional line ministries such as the Ministry of Health, Ministry of Women, and the Ministry of Education. The ministry would design the programs based on rigorous evidence. The central, regional, and local governments, as well as local communities, non-profits and businesses would act as partners in the implementation of the program. Cuna Más would launch a home visiting program and scale up nationally. In early 2012, Cuna Más sought help from Sally Grantham-McGregor. McGregor worked alongside others who had been involved in adapting the Jamaica model in Colombia as well as Peruvian early childhood experts to create proto- cols for the home visits. The team spent months adapting the Jamaica curriculum and materials to the Peruvian context, creating different materials for both the Andean and Amazon jungle contexts. They prepared training materials and identified local songs, games and play materials. Cuna Más used Peruvi- an artists to recreate illustrations that would reflect local lifestyles, geography, and customs. And the program outlined steps for the local teams to build toys and puzzles made of local materials. Home visits were to be carried out by specially trained community volunteers, who ideally had at least a high school education. Each community volunteer would be in charge of approximately 10 fami- lies. For about 10-12 hours of work each week a volunteer would receive approximately 360 Peruvian Soles (approximately 110 US Dollars). All volunteers would be trained in the curriculum for several days and would be taught how to interact with children and mothers in a sensitive and responsive manner. They would be taught best practices in demonstrating to parents how to engage with their children. The home visitors would be managed by supervisors in charge of training and monitoring. Each su- pervisor, ideally with professional training in early childhood development, would manage approximate- ly 10 volunteers. A small technical team would support the supervisors with technical and administrative help. The technical team and supervisors would be recruited by the central Cuna Más team in Lima with inputs from a locally selected committee of five volunteers tasked with managing the local delivery (see Exhibit E for Cuna Más administrative structure). The districts eligible to receive the home visiting service would be selected based on one or more of the following criteria:  areas where the incidence of poverty was greater than 50 percent;  areas that fell within the scope of intervention of the conditional cash transfer program, Juntos;  areas that had a rate of chronic malnutrition greater than 30 percent for children aged 0-5; and toring and evaluation. Independent observers argue that this part of MIDIS played a key role in some areas and fell short in others, for example in effectively coordinating across sectors and levels of government. Cuna Más 5 of 18 HKS Case 2115.0  d  areas where 50 percent of the population lived in a rural community. These criteria yielded 713 districts comprising nearly 32,000 rural communities. Problems Scaling Up The plan was to officially launch Cuna Más in the southern region of Ayacucho in March 2012. But the team working on the home visiting service had not finished adapting the curriculum or materials. Cuna Más also had considerable trouble delivering materials to Ayacucho. For political reasons Humala and others agreed to go ahead with the launch as scheduled despite knowing that the actual home visit- e ing program was not ready. By the end of 2013, however, the program had picked up speed. The home visiting service was im- plemented in 212 rural districts across the country. That number grew to 351 districts in 2014. The number of families benefitting from the home visiting service between 2013 and 2014 also grew more than 50 percent from 33,000 to nearly 52,000 children. By 2015, Cuna Más reached approximately 82,000 children through the home visiting service (in a total of 578 districts).6 Despite its impressive growth, a host of issues plagued Cuna Más, from lack of available staff to lo- gistical issues in reaching remote areas. In the early years the program was unable to quickly handle procurement processes such as purchasing goods (toys), reproducing material (curriculum or workbooks for staff), and making payments. This meant that implementation in places like Ayacucho began without materials (toys or curriculum). To make matters worse, an economic slump that began in 2014 forced the government to keep budgets from growing. Humala himself was besieged by political turmoil. Be- tween 2011 and 2015, Humala changed prime ministers seven times and replaced ministers of nearly all the major ministries, including MIDIS. At the same time there were new appointments made to vice ministries at MIDIS and the top leaders at Cuna Más changed as well. The head of the home visiting pro- gram was replaced. Jorge Fernández was the third director of Cuna Más in as many years. Frequent leadership changes took a toll on the program as priorities shifted with each new leader. Building Capacity In each province that Cuna Más entered, it routinely ran up against two related challenges: lack of available staff and home visitors, and high employee turnover rates. Although the program model did not specify that Cuna Más staff had to have knowledge of early childhood development, such qualifications were desirable. In reality, though, hiring staff was difficult. According to Regina Moromizato, founder of Red Innova a non-profit organization that facilitated coor- dination between municipalities, Cuna Más was never able to fully staff a technical team. “The theory d The Peru government defines a rural populated area as one that has less than 2,000 inhabitants or less than 400 homes. e There was also symbolic value in starting in Ayacucho because of its long history of underdevelopment. It is one of the poorest regions of the country, set high in the Andes, and was the birthplace of the Shining Path insurgency decades earlier. Cuna Más 6 of 18 HKS Case 2115.0 focuses on developing capacities of children,” she said. “So Cuna Más would like professional staff to have a background in early education. But according to the Ministry of Education, in 2015, Peru had a deficit of 40,000 preschool teachers. The Ministry of Education, which itself had an ambitious program for the same age range couldn’t cover all the positions, even though they offered better pay and work- ing conditions than Cuna Más. That meant that Cuna Más couldn’t attract professionals with that profile, much less in rural areas where, in addition to having technical training, they needed to speak the local language.”7 In running up against this barrier, Cuna Más expanded both its list of desirable candidates and the training it provided to technical staff and home visitors. Jorge Luis Fernández, executive director of Cuna Más, said that over time, the program learned to make adjustments on the fly. “We adapt the job pro- files to whatever is available among staff candidates in the area,” he said. “If we are looking for health professionals, for example, and there are fewer in one area we try to adapt the job description.”8 This flexibility, while helpful when recruiting, led to inconsistencies in the way the program was rolled out in different areas. The most important element of the Cuna Más model—the home visitors—were often the most diffi- cult to find, hire and retain. Under the program guidelines it was not compulsory that home visitors be educated at the high school level, even though such a level was desirable. But in many of the rural areas, volunteers, with or without secondary education, either did not exist or were otherwise employed. “In the 1970s, 80s and even the 90s, you had a lot of community volunteers that were very interested in participating in these kinds of programs,” explained Moromizato. “But Peru has grown a lot economical- ly since then. When mining, gas and oil companies started investing in rural areas, the economy got acti- vated. Now these companies hire local people and they earn a good salary. This means that working as a home visitor for a very small amount of money is not very attractive to many people. They can clean hotel rooms for a lot more money.” Cuna Más worked with networks of community leaders in each district to identify potential volun- teers. In some instances, women who were trusted in the community, but who did not have the level of education the program desired were hired and then trained by the technical teams. Even when Cuna Más was able to hire volunteers, it had trouble retaining them. Moromizato argued that the Cuna Más program requirements while sound in theory were, in practice, not positioned to evolve with the realities on the ground. She noted that the nature of the Peruvian economy meant that people in areas of high poverty often migrated for jobs, particularly high school graduates, who had higher hopes for their future. “Cuna Más expects results from community volunteers,” Moromizato said. “That’s not realistic. The program has a high rate of turnover (not just in terms of volunteers, but also in terms of its professional staff) and everyone knows that the success of any program is continuity. If you're changing your personnel every three months, how do you build something that has continuity? How do you create strong links with the families and communities? If every few months you have to do capacity building, or training, it's like there's a hole in this system.” Cuna Más 7 of 18 HKS Case 2115.0  Marta Rubio and Caridad Araujo of the Inter American Development Bank pointed out that on many occasions money for payments arrived late to the communities. “Anecdotal evidence suggests that a lot of staff turnover was due to delays in payments,” they said.9 Over time, Cuna Más’ supervisory staff found that administrative demands, such as making pay- ments and filing paperwork detracted from their ability to monitor and mentor community volunteers during their home visits. Quality vs. Scaling Up From the beginning, Cuna Más teams across the country had to strike the difficult balance between scaling up services and ensuring quality. Research suggested that in the 2000s, many developing coun- tries that tried to improve access to primary education ended up sacrificing the quality of the education delivered. The Cuna Más leadership too realized that with the emphasis on growth, quality of program delivery was being ignored. By 2014, following several changes in nearly all the top leadership positions, the program began to shift the focus from scaling up the program to implementing it better in places where it was already operating. Governance Mechanisms Several civil society groups involved with Cuna Más believed the program’s centralized structure was the primary reason behind Cuna Más operational challenges. They argued that the Ministry of Develop- ment and Social Inclusion by managing the program through a central team at the national level was failing to build the capacity of municipalities, which were in a better position to understand local needs. Decentralization: Fernández, who was often on the receiving end of this criticism believed it was a “problem of perception, rather than a real issue.” According to him the numbers spoke for themselves. Of Cuna Más’ total staff of 1,800, less than 10 percent—160—were located in Lima’s central headquar- ters. “This team basically is the administrative part of the program,” Fernández explained. “They do pro- gram design, supervise technical issues, legal issues, information technology, etc. The rest are in the field, in the various provinces working directly with community actors.” When we go into the district the first thing we do is communicate with the municipality. The municipality helps us identify the target population: areas where the children are in need and families are living in poverty. We get in touch with whatever organizations are in the community. We have a public meeting where we explain what the program is, what our intervention is going to be, what our program delivers. Once the community accepts the proposal they themselves select a committee of five people that will run the program. And this committee receives the funds from Cuna Más to acquire material that will be needed for the intervention. This committee also selects potential facilitators, caretakers, and social workers and pays them. Caridad Arajuo and Marta Rubio at the Inter American Development Bank also supported the view that Cuna Más was adequately decentralized. To them the problem with scaling up tied back to the issue Cuna Más 8 of 18 HKS Case 2115.0 of lack of capacity. “Cuna Más had a very small team in Lima to implement a program that was expand- ing too fast,” Araujo and Rubio noted. At the local level the administrative structure rested on the com- mittee of leaders who volunteered to carry out management duties. “Similarly, at the regional level, the program is managed by regional units that receive guidance and support from Lima but also function quite autonomously,” Araujo and Rubio said (see Exhibit D for Cuna Más administrative structure). They believed a stronger presence in Lima might have helped ensure higher quality and a more homogeneous implementation of the program across the country. “Certainly, the Lima team would have had to take into account the local realities and needs and work together with the regional teams to adjust, adapt and contextualize the model as much as possible,” they said. “Admittedly, this type of adaptation did not happen enough. We believe this was in part due to the small size of the team in Lima, which had many more responsibilities than they could possibly handle. These teams did not have the technical ca- pabilities nor did they receive the required training to support and monitor the implementation.” For Cuna Más’ partner organizations, the bigger concern around decentralization was that Lima con- trolled the purse strings, as well as the decisions about where to intervene and how. Leonardo Yanez at the Bernard van Leer Foundation and others sought a scale-up strategy that engaged regional and local municipalities as equal partners. Yanez cited the example of another early childhood program, Familia Feliz (translated: happy family), created in 2012 by the governor of Piura, a northwestern province in Peru.10 Although Familia Feliz was born independently of Cuna Más, it had striking similarities. One of Familia Feliz’s core programs was much like Cuna Más’ home visiting service, but, by contrast, it had been created by the regional government and was co-funded by local districts. It was embedded in a broader regional strategy on early childhood that also included efforts to reduce malnourishment and improve health outcomes. Between 2012 and 2015, Familia Feliz served 23 of Piura’s 65 districts, com- pared to Cuna Más’ 10 districts. Although the two programs served different areas, some saw Familia Feliz’s model as more integrated. “Knowledge of the place is very important in terms of public policy, both in terms of scalability and in terms of impact,” said Tomás Flores Noriega, the regional manager for Social Development in Piura in charge of Familia Feliz. “One of the key advantages of Familia Feliz is that, both in its design and in its implementation, there is a political and budgetary commitment from region- al institutions. Peru is a really complex country. If a social program is not working through the regional or local government, many times, there's more resistance on the part of the various stakeholders or a slower implementation process.”11 Value of a Multisectoral Approach: Jorge Fernández, however, underscored that in Peru, programs like Familia Feliz were the exception rather than the rule. Most social programs and policies tended to be ad-hoc in nature and varied widely among the regions. This ad-hoc approach was precisely what Cu- na Más was trying to address with its multisectoral approach, Fernandez explained. As one of several programs in the national government’s overall strategy on early childhood development, Cuna Más was mandated to integrate policies with multiple ministries, at different levels of government, such as health, education, nutrition, and child protection. “Each government program has its own specific set of tasks,” said Norma Vidal, Deputy Minister at MIDIS. “But if we’re intervening in the same homes, our interventions must be well coordinated.”12 Cuna Más 9 of 18 HKS Case 2115.0  Cuna Más adopted a case-by-case approach when entering a province or municipality to ensure that it worked in concert with existing capacities and actors. In communities where there was not much of a government presence in terms of early childhood, Cuna Más worked directly with municipalities and civil society. And in instances where there was a robust early childhood program already underway, as in Familia Feliz, Cuna Más coordinated efforts with the various entities and ensured there was no duplica- tion of services. Incentives and Tradeoffs: Fernández claimed that the process of customizing the program for each target area took time, but it also ensured the long-term sustainability of the intervention. He recalled a tense negotiation in the province of Apurimac (southern Peru) where the regional government ran a program that also used health volunteers to conduct home visits to help reduce anemia and malnour- ishment. “The program was getting good results,” said Fernández. “They weren't interested in Cuna Más coming to the area because we would distort and maybe even taint the progress they had made. Our home visitors get a stipend in recognition of their work. But their health volunteers get no money. If we went in, we would have introduced a perverse stimulus into the situation.” In hindsight, Fernández ad- mitted that Cuna Más should have identified this conflict of interest as a potential barrier before trying to start operations in Apurimac. Later, though, the national and provincial governments agreed to work in different districts. “Together we coordinated our messaging and our actions,” said Fernández. “So we could reach the goals that we both have.” The larger issue at the heart of the debate was how best to increase the capacity of local govern- ments to plan, budget and implement Cuna Más. Years of decentralization efforts in Peru had yielded mixed results. Staff at the sub-national levels often needed professional development in order to im- plement social programs. This same lack of capacity, combined with a national push to move funding to a results-based framework, meant that local governments often had to prioritize the programs they fo- cused on. It was more efficient, for example, to use limited resources and staff time to deliver nutrition or health programs that had clear, measurable goals, than to focus on a home visiting program where the outcomes were hard to quantify. To implement at scale with quality, Cuna Más needed to align the incentives of national leaders with local governance structures, as well as planning and budgeting mech- anisms. But it was unclear if and when such an alignment would happen. Adapting to Local Contexts Closely tied to the issue of effective governance was the concern that Cuna Más was operating in a geographically and ethnically diverse country with varying levels development. Cuna Más partner organ- izations including Salgalu and Red Innova argued that the program did not adequately take into account the many differences between communities. “For Cuna Más to stop being an experiment and start being a large-scale program,” said Salvador Herencia of Salgalu,“we have to be realistic about the local environment.”13 Most of the areas that Cuna Más was trying to reach were high poverty areas in remote parts of the country that often lacked basic services such as schools and health centers. Apart from the lack of human capacity to implement the Cuna Más 10 of 18 HKS Case 2115.0 program, most residents of such areas did not see value in early childhood education. “That means we have to bring awareness and create ownership of the program in society,” said Herencia. On one level, Cuna Más had to build ownership for the program among local political and community leaders. “But in 2015, only 25 percent of mayors even knew that Cuna Más public investment funds existed or that they could use it,” explained Herencia. On another level, Cuna Más had to work with recipients of the pro- gram to make the linkage between early childhood education and long-term reduction in poverty. Moreover, the model with its emphasis on psycho-social stimulation had not accounted for the need in local communities to address entrenched issues of anemia and malnourishment. “Cuna Más was look- ing at learning improvements, as illustrated by the goals listed in their budget, which are focused on mo- tor and cognitive skills, language, and social and emotional development, “ said Regina Moromizato of Red Innova. “But they weren’t taking into account the other needs of the areas they were serving— areas with extreme poverty. Issues like anemia or malnutrition during the early years are a key concern in the minds of decision-makers and community leaders in those areas. If you didn’t solve those issues, how could you persuade the population to prioritize early childhood learning?” Under Fernández, Cuna Más began moving towards a stronger focus on nutrition, hygiene, health, and going to well-child visits, but the curriculum itself did not include these elements and the program relied on home visitors to deliver the information during home visits. “The original Jamaica curriculum did not have a nutritional counseling component,” explained Caridad Araujo and Marta Rubio of the In- ter American Development Bank. “Cuna Más added it into the structure of the home visit. However, there was less effort invested in developing curriculum for this part of the visit, little coordination with the community health workers responsible for nutritional counseling in rural Peru, and little training of program staff in this area.” Moreover, Cuna Más faced logistical issues in getting toys and puzzles to remote parts of the coun- try. In the early days of the program, the central staff at Cuna Más hoped that communities would en- gage in manufacturing their own toys and puzzles, thereby sidestepping potential bureaucratic delays. But when concerns about the quality of materials used in different sites emerged, the team in Lima de- cided to manufacture the materials centrally and distribute them to each community. But some of the most marginalized Peruvian communities lived in the Andes highlands, where geographic isolation made it difficult for Cuna Más to deliver materials. With no buses or paved roads, local staff and volunteers were required to walk long distances, carrying materials. Similarly, Cuna Más ran into issues when adapting curriculum materials to the myriad indigenous languages and cultures in Peru “We have a policy of adapting all our children’s material to fit in with lo- cal cultures,” said Fernández. “In rural areas, all interventions are done in the native languages.” This policy, however, placed further burdens on overextended Cuna Más teams. In the Amazon jungle, for instance, the father’s role as a major figure in the family ran counter to the culture of other communities in the country where mothers played a dominant role in children’s development. Cuna Más took the time to adapt its materials to ensure that community volunteers in the Amazon incorporated fathers into the mix. “We really consider the feedback from all the different areas,” explained Fernández. But Cuna Más 11 of 18 HKS Case 2115.0 Herencia questioned the efficiency of a program that did all the planning in Lima. “There’s a big a need to generate the capacity to implement the program on the local level and to successfully decentralize early childhood services, be it Cuna Más or another proven model,” he said. “Otherwise, it is like bring- ing a Ferrari to the jungle that uses very expensive fuel and can only be driven by a driver who’s not in the jungle,” said Herencia. Measuring Impacts Proof of Cuna Más’ success, and by extension its longevity, ultimately depended on the outcomes it generated. Internally, Cuna Más monitored the quality of the services it provided through user satisfac- tion surveys. In addition, Cuna Más checked to see if and when children reached certain cognitive, phys- ical, nutritional and emotional developmental milestones. “If we have children who reach the expected level of development, and if the users are satisfied,” said Fernández, “then we can say that we’re ful- filling our remit by providing a quality service.” According to Caridad Araujo and Marta Rubio of the Inter American Development Bank, data availa- bility for the home visiting service was a barrier to effective monitoring. The program “failed to build an administrative data system that could produce basic information about its service efficiently and on a timely basis,” they said. “It has taken a lot of time and resources to fix this over time and still the pro- gram administrative data are likely to have quality problems.” Cuna Más nonetheless found the monitoring efforts useful in making necessary program tweaks. “We monitor the process of implementation to improve Cuna Más,” he said. “We originally planned to make home visits once a week in all cases. But for families with difficulties we are considering increasing the frequency. The infrastructure models for the childcare centers have been gradually adapted to suit each region as well, and we have been increasingly diversifying provision.” But Fernandez acknowledged that more independent research would be needed to examine the quality of the services Cuna Más pro- vided. To single out the direct impacts of Cuna Más, the government recruited the Inter American Devel- opment Bank to conduct an impact evaluation of the program. The results would not be available until after the 2016 elections. But Regina Moromizato at Red Innova hoped for more. “Those of us who've spent some time in the field think there's a lot that needs to improve in the program,” she said. According to the Peruvian constitution, President Humala was not eligible to run for a consecutive term. Even as observers of Peru’s politics argued about the candidate most likely to win the election, they agreed that a national focus on early childhood development was here to stay. Whether the belea- guered Cuna Más program would remain the same or morph into a different program, under a different name, only time would tell. Cuna Más 12 of 18 HKS Case 2115.0 Appendix 1: Comparison of Early Child Programs in Colombia and Peru Notes: * Colombia’s home visiting program was a pilot effort and not an integrated part of the government’s overall early childhood development strategy De Cero a Siempre (from Zero to Forever). ** The home visiting pilot in Colombia ran for 18 months between 2010 and 2011. Cuna Más 13 of 18 HKS Case 2115.0 Exhibit A: Peru in Detail Source: Created by Elizabeth Moran. Maps adapted from freevectormaps.com. Cuna Más 14 of 18 HKS Case 2115.0 Exhibit B: Peru at a Glance Peru Poverty and Economic Growth 2004-2012 Demographic Statistics Total Adult Total Population Population Under 5 Urbanized Population Literacy Rate (2013) (2013) (2013) (2009-2013) 30,376,000 2,924,000 94% 78% Source: Chart adapted from PWC, “Mining Industry: Doing business in Peru,” 2013. Table data from UNICEF Statistics. Cuna Más 15 of 18 HKS Case 2115.0 9 0 –2 0 13 gets –2 0 13 Exhibit 9 0relating F i gu r e 3 : Prevalence of chronic child undernutrition in C: Peru Early Childhood Statistics nificant children under 5 by residential area 5A. rgetsThe relating F i gu r e 3 : Prevalence of chronic child undernutrition in utrition fell gnificant Malnutrition Rates children under 5 by residential area 905A.and The %utrition set forfell 90 and % set for n the nt e situations ion rates in in the n, Lima and ent situations ectively, tion rates in elica, n, Lima and Source: INEI.21, 22 %, 35.6% ectively, velica, Source: INEI.21, 22 %, 35.6% chieved the F i gu r e 4 : MDG 4 indicators for children under 5 5 child Child and Infant Mortality 1000 live achieved the F i gu r e 4 : MDG 4 indicators for children under 5 duction er 5 child erperiod. 1000 live 1-1992 eduction ss,period. and the 1 deaths 91-1992 urred s, andfor the 015 of 21 deaths 18 rpassed curred forin deaths 2015 of 18 urpassed in 6 deaths Source: INEI.21, 22. gradually the target 21, 22 Underweight Prevalence in children under 5 Source: INEI. Primary School Net Enrollment o000 live gradually births, which reduced to 185 in 2000 and 93 in 2011. significant h the target During this period, the number of births attended by Urban Rural Urban Rural 0a.000 In 1996, live healthcare professionals births, which reduced toin185 a health institution in 2000 and 93andin 2011. 000 live significant coverage of antenatal During this care period, the increased number significantly. of births attended by 2% 6% 97% 98% ea. In 1996, healthcare professionals in a health institution and 0 000 live coverage of antenatal care increased significantly. 13 13 Source: Charts from World Health Organization, “Success factors for women’s and children’s health: Peru,” 2015, Figures 3 and 4 on page 13, http://www.who.int/pmnch/knowledge/publications/successfactors/en/; Table data from UNICEF Statistics. Cuna Más 16 of 18 HKS Case 2115.0 Exhibit D: How Cuna Más is Implemented Cuna Más 17 of 18 HKS Case 2115.0 Endnotes 1 Data provided by Cuna Más. 2 World Bank Economic Indicators. 3 Ibid. 4 Ibid. 5 This and other quotes by Sally McGregor are from an interview with the World Bank: Q/A: Sally McGregor on Jamaica’s Pio- neering Role in Early Childhood Development, available http://blogs.worldbank.org/education/qa-sally-mcgregor-jamaica-s- pioneering-role-early-childhood-development (accessed January 30, 2016). 6 Data provided by Cuna Más 7 Unless otherwise stated all quotes attributed to Regina Moromizato were drawn from an interview with the author on Febru- ary 16, 2016. 8 Unless otherwise stated all quotes attributed to Jorge Luis Fernandez were drawn from an interview with the author on Feb- ruary 17, 2016. 9 Observations and quotes attributed to Caridad Araujo and Marta Rubio are based on email correspondence with the author on April 18, 2016. 10 Unless otherwise stated all quotes attributed to Leonardo Yanez were drawn from an interview with the author and Julie Boatright Wilson on January 18, 2016. 11 Unless otherwise stated all quotes attributed to Tomas Flores Noriega were drawn from an interview with the author on Feb- ruary 8, 2016. 12 Early Childhood Matters, “Key Success factors are joint management and intersectoral cooperation,” an interview with Jorge Luis Fernandez and Norma Vidal, available http://earlychildhoodmagazine.org/key-success-factors-are-joint-management-and- intersectoral-%C2%ADcooperation/ (accessed February 22, 2016). 13 Unless otherwise states all quotes attributed to Salvador Herencia were drawn from an interview with the author on Febru- ary 2, 2016. Cuna Más 18 of 18 HKS Case 2115.0