How did Peru implement?

In 2006, at the urging of a united civil society campaign, every presidential candidate signed a pledge to reduce chronic malnutrition by five percent for children under five years old over the next five years. The 5x5x5 campaign leveraged the 2006 presidential election to elevate the issue and ensure that stunting became a top priority not only for the winning candidate, but for political parties across the spectrum going forward.

Upon his election in 2006, President Alan Garcia went a step further and pledged to reduce chronic child malnutrition by nine percent. President Garcia asked the Prime Minister’s office to take the lead along with the Ministry of Economy and Finance. Such high-profile political backing gave the stunting reduction programs instant prominence.

Regional leaders also became champions for stunting reduction when they, as part of the EUROPAN incentive program and later the Peruvian-funded Fondo de Estimulo al Desempeño y Logro de Resultados Sociales (FED) program, publicly pledged to meet ambitious goals

After 2011, national stunting reduction efforts were led by new ministry focused exclusively on social inclusion, the Ministry of Development and Social Inclusion (MIDIS).

In 2007, Peru launched a rigorous review and rationalization of its existing poverty alleviation and nutrition programs. Academics, civil society leaders, and government officials collaborated to determine which of the country’s 80-plus programs were evidence-based and cost-effective. Those supported by the evidence were streamlined for even greater efficiency and incorporated into a new data-driven national strategy, Crecer.

The new strategy proved a decisive shift for the country. Peru’s uncoordinated and ineffectual food supplementation programs were deprioritized. In their place, roughly two dozen tightly interwoven and mutually supportive, data-driven initiatives were elevated. The initiatives, and the new strategy as a whole, focused more narrowly on connecting the target demographic (pregnant women and children under three years old) with preventive health services; improving the quality and availability of those services in poor areas; and creating a health-seeking culture in the target demographic.

The new strategy streamlined Peru’s approach on three levels: the range of programs Peru supported was much reduced, the target beneficiaries more limited, and the programs themselves were more focused, featuring only a few key activities.

Each of Peru’s stunting reduction programs was geographically and demographically targeted to ensure it reached the most at-risk pregnant women and children in their first 1,000 days living in the poorest communities in the poorest states. Peru’s experience with targeting has been mixed. Each program used a slightly different formula for identifying beneficiaries with varying results.

Crecer

The government’s national anti-poverty strategy, Crecer, was initially piloted in 200 of the poorest districts in the country (out of a total of over 1,800 districts). Later, this was expanded to 880 districts in the country. These districts, which covered nearly half the country, had poverty rates of more than 50 percent, illiteracy rates of more than 30 percent, and malnutrition rates of more than 30 percent. Eventually, the strategy was scaled to 14 of the country’s states, out of 26.

Juntos

The conditional cash transfer program, Juntos, was initially piloted in Ayacucho, one of the poorest states in the country and the epicenter of the Shining Path guerilla war in the 1980s and 1990s. Juntos was later expanded to families in about 100 of the poorest districts of the country (out of over 1,800). The prioritized districts were identified based on five criteria, which included levels of child malnutrition, unsatisfied basic needs (lack of water and sanitation), and the percent of households previously affected by terrorism during the Shining Path guerilla war. Within these districts, there was further targeting of families that scored below a certain threshold on a formula which included the following inputs: percent of illiterate women in the household, existence and number of home appliances, and access to public services such as clean water, electricity, and sanitation. Lists of approved families were further screened by local community members. This last step may have corrected errors produced by the formula, but also may have added a layer of subjectivity that could introduce bias. Families were re-evaluated every four years to determine if they still fell within the income guidelines of the program. While Juntos eventually expanded geographically to cover nearly the entire country, it continues to have a strictly targeted beneficiary population.

SIS

The health insurance program, Seguro Integral de Salud (SIS), used an undisclosed proxy means test, administered at health facilities, to identify potential beneficiaries. Applicants provided information about demographics, education, occupation, assets, and housing conditions. This information was plugged into an algorithm, kept secret to reduce the risk that patients or providers would game the system, to determine eligibility. Enrollment was immediate if there was a computer at the facility but took up to eight weeks if there was not, which was more common in poorer areas with more SIS recipients. Despite the secrecy, means testing at the facility was vulnerable to gaming. Non-poor people in bad health were incentivized to lie to try to obtain coverage. Given the inadequate funding for SIS and therefore its spotty reimbursement record, providers were also incentivized to limit the number of recipients. Estimates of the program’s success at targeting vary. But overcoverage of the non-poor was at least 12 percent, and undercoverage of the poor was at least 16 percent—and probably almost twice that. These inaccuracies reflect not only strategic behavior on the part of both patients and providers but also the weakness of a proxy means test. In 2004, Peru formally developed a more accurate household targeting system, but the rollout has been extremely slow; implementation is still limited to pilots in and around Lima and other urban centers.

The World Bank and the Initiative against Chronic Infant Malnutrition (IDI) helped Peru establish “results-based financing” programming in 2008.

Results-based financing served two purposes. It ensured that Peru’s regions received funding commensurate with what they were being asked to achieve, instead of funding based on past benchmarks. Under results-based financing, poor regions with high rates of stunting received large increases to bring their funding in line with realistic expectations of the costs that would be incurred to meet their goals. This change meant that poor regions, long accustomed to underfunding, suddenly had the resources to hire the staff they needed and buy the equipment necessary to serve the large numbers of poor families they were being asked to serve.

Results-based financing also served to protect those allocations from being used for other needs. Under results-based financing, funding allocations could only be used for their intended purpose. For example, funding for vaccines in a remote clinic could not be appropriated to cover office expenses in a regional capital. Results-based financing is now common in many low-income countries.

Peru’s results-based financing programs included Programa Articulado Nutricional (PAN) and Programa Salud Materno y Neonatal (PSMN).

To ensure effectiveness and increase accountability, Peru expanded its national survey efforts and built monitoring mechanisms into each of its programs.

Data
The Demographic and Health Surveys (DHS) launched in 1986 and collected data every three to five years until it became an annual survey starting in 2007. Four years later, Peru increased the sample size to allow the government to better monitor programmatic impact and health care usage. This annual survey, which includes intervention coverage and stunting outcome, is used for both measurement and evaluation, as well as accountability.

PAN
The results-based-financing program PAN required each health center to report data on staffing and supplies down to the number of syringes used and number remaining in stock. This data is maintained on the Sistema Integrado de Gestión Administrativa database (SIGA), which has been installed in 158 health management units across the country and allows officials to track how many vials of a given vaccine, syringes, and other supplies are in hundreds of clinics.

Juntos
The conditional cash transfer program, Juntos, required local managers to report on each child’s school attendance rates as well as the health services provided to each child and mother in their service area every two months. This information was entered into a national database, allowing government officials to determine where children were receiving the required health care and where they were not. Parents enrolled in Juntos were required to obtain national identity cards for their children, allowing authorities to track the health care each child received no matter where they moved

Monitoring expenses
The Ministry of Finance maintained an online budgeting platform, Sistema Integrado de Administracion Financiera, (the Integrated Financial Management System, SIAF), allowing the public to track allocations and expenditures through the online portal Transparencia Economica.

Civil Society
A unified and interested civil society also played a role in monitoring and holding the government accountable. Local governments held semiannual meetings with civil society and community leaders to share data on intervention coverage, quality of care, effectiveness, and impact. The civil society groups Iniciativa contra la Desnutricion Infantil and Mesa de Concertacion also convened quarterly or semiannual meetings, tracked impact, and regular progress reports.

In 2011, the European Union provided Peru with $72 million to pilot an incentive program (EUROPAN) to expedite progress towards critical milestones. At the outset, officials from the three poorest states, Huancavelica, Ayacucho, and Apurimac, publicly signed a pledge to meet ambitious stunting reduction targets. When they reached these targets, they received prestigious national prizes, were held up as national models, and their states received additional discretionary funding.

This program’s success led Peru to launch the domestically-funded incentive program Fondo de Estimulo al Desempeño y Logro de Resultados Sociales (FED) to expedite progress across all 26 states. FED worked in much the same way as the EUROPAN incentive pilot. It provided financial rewards and prestigious prizes such as the Sello Municipal (municipal seal) to states and municipalities meeting their targets. The Sello Municipal honors were distributed at an awards ceremony at Lima’s Gran Teatro, where more than 1,000 people attended.

Parents were also incentivized, through the conditional cash transfer program Juntos, to ensure their children received the recommended preventive health care.

Two types of technical support helped Peru achieve its stunting reduction goals. First, external organizations, such as UNICEF, CARE, and the World Bank, served as advisors in the design and development of key Peruvian poverty and stunting reduction programs. Secondly, the central government provided technical support to both regional and district governments for the implementation of those programs.

Support from external partners:

• The World Bank advised Peru on the establishment of its results-based financing programs and on a national communications campaign to ensure the issue of stunting was understood and interventions supported.
• The European Union advised Peru on the development of the EUROPAN incentive framework to expedite the implementation of PAN.
• USAID supported Peru’s expanded survey efforts early on and funded the UNICEF and CARE pilot programs which demonstrated that stunting reduction was possible.
• UNICEF, CARE, and more than a dozen other large non-profits and multilaterals provided guidance through the Initiative against Chronic Infant Malnutrition.

Support from the national government:

• The national government created technical support teams to provide guidance to regional and local governments and developed national guidelines and protocols for to ensure best practices for preventive care and counseling. Within the Ministry of Economy and Finance, for example, a team of experts with Ministry of Health backgrounds helped ensure that budgeting for the key health programs was in line with goals.
• Ministry of Development and Social Inclusion (MIDIS) established regional liaison teams across the country to help officials plan, implement, and coordinate strategy and programs.
• Traveling teams of experts. A group of approximately 100 doctors, nurses, and economists travelled throughout Peru to help local officials budget properly to ensure adequate resources for program implementation. Clinic staff received training from a travelling group of some 100 nurses and social workers equipped with guidelines to ensure maternal counseling, child and pregnant mother growth monitoring, and treatment followed best practices established by the World Health Organization.
• Guidelines: PSMN, for example, developed standards and technical guides (with allowances for appropriate cultural adaptation) for health care procedures for women and newborns. Health providers across the country participated in trainings to reinforce the messages and protocols in the national guidelines. Likewise, Juntos counselors who educated mothers about nutrition and handwashing were also encouraged to follow specific guidelines. PAN also distributed standards and protocols and made some available online.