April 13, 2016

Aligning Interests in a Decentralized Country: A Pay-for-Success Mechanism Made in Peru

Cusco, Peru
Cusco, Peru (Sascha Grabow/Wikimedia)

When talking about international development, it is often said that fixed recipes do not work well in every context. Pretending that, as an outsider, one could determine what should be done to fix or improve the provision of public services in a certain context based in what worked in another place is risky and can even be harmful. This was the dilemma that the Peruvian government faced after 2006, when the decentralization process was finally "completed."

The quick and painful process of decentralization in Peru

Eight years ago, Peru changed the way it operated as a state, decentralizing such that it now has 25 Regional Governments and more than 1,800 Local Governments, each one with a democratically elected authority and its own bureaucracy and priorities. The theory that supports decentralization argues that by making the government closer to people, it is better prepared to understand and attend to the needs of their own jurisdictions.

The process of decentralization in Peru started with the transfer of financial resources to subnational governments and it continued with the transfer of functions (as broad as "providing health and education services in the region," without stating what that really entails). However, transfer of capabilities was not a serious task during the process. Therefore, it was not surprising when scandals of corruption and misuse of public budget were broadly covered by the media. At some point, news outlets were covering how Regional and Local Governments were using public resources to build parks in memorial to the most random things you could imagine or to make the municipality building look spectacular in locales where there was not even access to electricity.

Meanwhile, by 2009, Peru was still ranking among the last countries in the list of reading comprehension and math scores (see PISA Ranking 2009), and more than four out of every 10 children in rural zones had stunted growth (see MIDIS report). Nonetheless, according to the law, the Regional Governments were now autonomous and must decide for themselves what to do with the resources they had. Then, the question for the executive branch was how to align all of these new players around certain priorities?

The concern around malnutrition and the rise of incentive-based budgeting

In 2008, there was a big effort to implement results-based budgeting, which in Peru took the form of a linkage between the budget and its purpose (e.g., pregnant women attending their monthly check-ups, children attending primary school). This effort certainly gave a better sense about what were the desirable outputs that the government should focus on rather than the traditional focus on inputs (e.g., acquisition of laptops). However, an effort for better classification of public budgeting was not enough to promote better performance in the provision of services. While Regional Governments can assign a lot of resources to certain outputs, it does not assure that money will be actually spent, let alone the quality of expenditure.

In this context, in 2009, the Ministry of Economy found room to implement a new way of incentivizing entities to move beyond a budgeting exercise and actually implementing the services required. This time, the budget was used as a tool to generate incentives for Regional Governments to work against child malnutrition. The cost efficient actions to fight child stunting were already stated in the context of results-based budgeting one year before, but making them work required another intervention. That is when the EUROPAN was created.

The devil is in the details: The need for micromanagement

In 2010, EUROPAN, an incentive-based budgeting (or pay-for-success) mechanism to reduce child stunting was implemented in three regions: Huancavelica, Apurimac, and Ayacucho, which are some of the poorest regions in the country. The mechanism consisted in setting a list of goals at two different levels: (i) management goals and (ii) provision of services goals that the Regional Government committed to fulfill and if so, it would receive a certain amount of additional resources in their budget.

The mechanism was designed to be implemented progressively, so the first indicators to be evaluated would be the ones that build capabilities and promote efficiency to provide services (such as planning, reporting and data management, stock availability of inputs, staff availability in health centers, etc.) and the second level of indicators assessed the coverage of the provision of services (such as percentage of children that attend their growth monitoring visits, percentage of children that have received micronutrient supplements, etc.). Additionally, full-time support from the Central Government was provided to the Regional Government to help them to comply with their goals.

Incentive-based budgeting applied in two phases to build capabilities in Regional Governments

The logic behind this mechanism is that building capabilities in the Regional Governments would help them to find their own ways to make things work in their own context. The mechanism did not tell every Regional Government the best way to comply with their targets but, instead, it only set the targets and left room for the Regional Government to determine its best strategy to meet them.

Four years later, according to the Demographic and Health Survey (ENDES 2009-2013), some indicators such as the attendance of children to growth monitoring visits and iron supplementation increased, especially for Huancavelica and Ayacucho (see Graph 2).

Graph 2: Coverage of services increased during the period of intervention, especially in Huancavelica and Ayacucho

Source: ENDES

Furthermore, during this period, Huancavelica reduced stunting rates from almost 54 percent to 42 percent; Ayacucho reduced it from 41 percent to 28 percent and Apurimac, from 35% to 29%.

Graph 3: Outcomes have improved in the three intervened regions

Source: ENDES

Surely, causality cannot be attributed to these numbers (a rigorous analysis needs to be conducted to determine it... researchers, work to be done here), but people do recognize how things changed once the mechanism was implemented. Actually, some testimonies argue that the value of the mechanism is not the offered money per se, but the power to align all stakeholders in a Regional Government to work towards clear objectives, from top decision-makers to the front line and even civil society.

Before this intervention, it was possible to find some isolated cases of good results of reducing malnutrition, but it was highly dependent on the particular motivation and commitment of some people working in the Regional Government and was not due to a systematic effort of all the stakeholders involved in the process. Moreover, civil society was barely involved and rarely asked for positive results on reducing malnutrition. However, now that there exists the possibility of losing extra budget if the Regional Government does not meet the goals, and that information is publicly available, civil society is empowered. On that note, one of the coordinators of the EUROPAN stated in an interview with the author* on February 2, 2016:

“The mechanism [EUROPAN] helped the Regional Governments to reveal information for accountability. In one of the regions, there was a huge movement in the media accusing the Regional Government they were about to lose money because of inefficiency.”

In 2014, the incentive-based budgeting experience was replicated in all 25 regions under the name of FED (acronyms in Spanish), but this time the objective has moved beyond reducing child stunting. Now, it includes a broader set of goals around promoting early childhood development (including nutrition, education, and access to sanitation and to government-issued identification). A much more ambitious mechanism is put in place and the results are yet to be assessed to determine if this could work for a broader set of purposes. I'll keep you posted.

* This interview was conducted in connection with thesis research by the author and Annie Chumpitaz, also a student at Harvard Kennedy School.

The views expressed in the Government Innovators Network blog are those of the individual author(s) and do not necessarily reflect those of the Ash Center for Democratic Governance and Innovation, the John F. Kennedy School of Government, or of Harvard University.

Related Topics

Related Topics