Digital government teaching case study

Digital Infrastructure to Prevent High School Dropouts

Manoela Vilela Araujo
2025

Introduction

I still remember one of the stories that pushed our team to move faster. João, a 16-year-old from a low-income neighborhood in northeastern Brazil, had recently dropped out of high school. He was a good student, but the pressure to help his mother with the household bills became too heavy. Between informal jobs and caring for his younger siblings, school quickly became secondary. João did not lack potential, but he did not have incentives, structure, and support. His case was not unique, and it reflected a broader pattern: thousands of students across the country were leaving school not because they didn’t value education, but because they couldn’t afford to stay.

Despite significant educational progress in recent decades, particularly the expansion of high school enrollment, school retention and completion remain a challenge in Brazil. Today, only seven in 10 young people complete high school by the age of 24, a rate that lags behind neighboring countries such as Argentina, Chile, and Peru. Among low-income youth, this number is even lower. According to data from the Continuous National Household Sample Survey (PNADC), only about 50% of young people from the poorest 20% of the population finish high school, compared to 92% among the richest. Racial inequalities deepen the problem even more. Evidence shows that Black youth aged 20 to 24 are 50% more likely to not graduate high school than their white peers (Pereira, 2022).

Far from being an individual choice, dropping out is often a structural response to poverty, inequality, and the urgent need to contribute economically to the family. Many students leave school not because they lack motivation, but because they must take informal and low-paying jobs to support their households or care for younger siblings while their parents work.

The consequences of school dropouts are well documented (Paes de Barros et al., 2021; Pereira, 2016). Evidence says that young people who do not complete high school face reduced opportunities in both formal and informal job markets, as well as lower lifetime earnings, and greater vulnerability to precarious health conditions, physically and mentally. From a national perspective, the economic cost of dropouts is estimated in billions of reais (Brazilian currency) annually, representing lost productivity, diminished human capital, and long-term pressure on public budgets, due to violence, drug trafficking and early pregnancy. According to PNADC data, when comparing individuals of the same characteristics (race, gender, age, and place of residence), those who complete high school but do not pursue higher education still earn, on average, 28% more over their lifetimes than those who do not finish high school.

In response to this multifaceted and complex challenge, Brazil's Ministry of Education launched the Pé-de-Meia (which translates to "Piggy Bank" in Portuguese) program in 2024. The program offers monthly and annual financial incentives to low-income students in public high schools, conditional on enrollment, attendance, academic progression and participation in the ENEM (Brazil’s national high school exam, similar to the SAT in the United States). But while its primary goal is educational, the program quickly became a case of digital transformation. To reach nearly four million students across all Brazilian states and ensure that payments were made securely and fairly, the government had to build a national digital infrastructure, integrating data from school systems and social registries.

This case study explores how an educational policy evolved into a digital governance transformation. It demonstrates how technology, data integration, and user-centred design can help reshape social service delivery and ensure access to rights for historically underserved populations. More than a cash transfer program, Pé-de-Meia represents a new model for building digital public infrastructure that supports inclusion and long-term human development.

The Public Servant’s Dilemma 

As a public manager at Brazil’s Ministry of Education, I was assigned to help lead the design and implementation of Pé-de-Meia. From the outset, it was clear that this would be more than a policy rollout; it would be a test of whether the federal government could deliver a citizen-centred policy at scale, using digital tools to reach some of the country’s most vulnerable youth. The Pé-de-Meia team was composed of public servants with different backgrounds and education policy consultants within the Ministry of Education. Back then, we had no in-house professionals with IT or systems development skills. While our team’s profile was not a limitation in itself, it quickly became clear that we would need to partner with a federal university or a specialized public agency to support us with data integration, system design, and digital infrastructure. 

Pé-de-Meia is a national educational savings program designed to democratize access to high school, promote student retention, and reduce dropout rates among low-income students enrolled in public schools. To achieve these goals, the program offers monthly and annual financial incentives based on four key milestones:

  • Enrollment: A one-time annual payment of R$200 (~USD $34) for each year of enrollment.

  • Attendance: R$200 (~USD $34) per month for up to nine months, conditional on a minimum 80% monthly attendance rate.

  • Academic progression: R$1,000 (~USD $171) at the end of each school year, accessible only upon completion of high school.

  • ENEM Participation: A one-time payment of R$200 (~USD $34) for attending Brazil’s national high school exam (ENEM, similar to the SAT in the United States).

The incentive design draws from behavioral economics, particularly the concept of loss aversion (Kahneman & Tversky, 1979), which suggests that people are more motivated to avoid losing a benefit they believe they already possess. By depositing progression bonuses annually but conditioning their access on graduation, the program encourages students to remain in school through the end of high school.

We knew the success of the program would hinge on our ability to solve a series of complex implementation challenges, such as: (1) integrating fragmented educational data from across the country; (2) ensuring monthly attendance tracking in real time; (3) establishing secure and accessible financial transfers for underage students; and (4) building direct communication channels with beneficiaries, many of whom had limited digital access or literacy. Many of these challenges had no perfect solution, and required us to balance scale, equity, legal constraints, and technological feasibility in every decision we made. Each of these challenges  will be detailed below, including the path we chose to address them.

1. Identifying the Right Beneficiaries

One of our earliest and most critical decisions was defining how we would identify which students were eligible to receive the Pé-de-Meia benefit. The law was clear: the program was intended for low-income students enrolled in public high schools across the country. But turning that legal criteria into an operational filter, applicable to millions of students, was anything but simple.

We considered creating a new national registry in which students or their families could self-declare income and school information, submitting documents directly to the Ministry of Education. While this approach offered autonomy and responsiveness, it would have required a massive effort in system design, document verification, and fraud prevention. It also risked excluding those we wanted to reach, students who lack the time, resources, or digital access to navigate bureaucratic processes.

The second option was to rely on Cadastro Único para Programas Sociais do Governo Federal (CadÚnico), Brazil’s unified social registry, which already includes socioeconomic information on families enrolled in other federal programs like Bolsa Família. We would then cross-reference that data with school enrollment records provided by state and municipal education departments. This option offered a faster path to scale, using pre-existing infrastructure. But it came with a major trade-off: CadÚnico data was not originally designed for education targeting, and the school data submitted by state and municipal education departments was fragmented, inconsistent, and lacked standardization.

We decided to move forward with data integration between CadÚnico and the education systems, with enrollment, attendance, and conclusion records. Although this option allowed us to operationalize the program quickly, and target students more accurately, it required a significant political and technical effort to align state and local education departments around a shared standard for reporting enrollment data. The decision also placed heavy demands on our digital infrastructure team, which had to create validation protocols, manage interoperability, and troubleshoot gaps in real time. This choice wasn’t the easiest, but it was the most viable. It demonstrated how existing data infrastructure, even if imperfect, could be leveraged to deliver targeted rights at scale if supported by strong governance and coordination.

2. Monitoring Monthly Attendance

Another critical requirement of Pé-de-Meia was the ability to verify student attendance on a monthly basis. The law established that students needed to attend at least 80% of school hours to remain eligible for the benefit. But in a country with more than 170,000 schools and deep regional disparities in digital infrastructure, how could we ensure consistent and timely attendance monitoring?

Our first option was to leave the task to schools and local education authorities, allowing them to report attendance data through their own systems and formats. This approach respected the decentralized nature of Brazil’s education system and avoided imposing additional burdens on state and municipal governments. However, it posed significant risks: the data would be inconsistent, slow to process, and prone to errors or omissions, especially in regions with weaker administrative capacity.

Alternatively, we explored developing a national system to collect and standardize attendance data. This solution enabled real-time processing and greater transparency, but required technical infrastructure, political buy-in from subnational governments, and a rapid implementation timeline. We ultimately partnered with a federal university to adapt and expand an existing platform, the Sistema Gestão Presente (SGP), into an educational hub.

With the financial incentive offered directly to high school students, we knew we had a compelling reason to engage state education departments. For the first time ever, they were being called to systematically collect and organize a wide range of student data, including enrollment, monthly attendance, academic progression, and participation in ENEM. This solution offered many benefits: it improved data quality, facilitated oversight, and created a foundation for long-term digital transformation in education. Many state education departments lacked prior experience with digital attendance tracking, and initial resistance had to be addressed through training, technical support, and sometimes, negotiation. Still, we recognized that investing in this system, as well as providing local public servants the ability to blend traditional skills with digital ones, would not only support Pé-de-Meia but also lay the groundwork for future educational data governance at scale.

3. Communicating with Students and Families

Pé-de-Meia's success would depend not just on eligibility but also on students’ knowledge of their right to access this policy. If students didn’t know about the program or didn’t understand why they were receiving (or not receiving) their benefit, the entire logic of incentives would be questioned.

We could have relied solely on traditional communication channels (schools, state and municipal secretariats, public campaigns), but these were slow, uneven, and often failed to reach students directly, especially in remote or underserved areas. We needed a way to speak directly to the beneficiaries.

We chose to integrate Pé-de-Meia into Jornada do Estudante (Student Journey), a federal government app already used for education services. This would allow us to send personalized updates to each student and offer a centralized platform for information about eligibility, compliance, and payment status. 

Moving forward with the app integration represented a strategic decision to treat communication as infrastructure, not just a side activity. However, challenges quickly started to emerge, because many low-income families lacked digital literacy, consistent internet access, or knowledge about how to use the app. 

Although the Jornada do Estudante app has now been downloaded by over 14 million users, there is  room for improvement, especially when it comes to reaching Pé-de-Meia beneficiaries. Approximately 40% of eligible students have yet to access the app, resulting in a lack of awareness about the payments and benefits they are entitled to. This remains one of the most pressing digital challenges that the federal government still has to address.

4. Authorizing Access to Bank Accounts

Once eligibility and communication with students were settled, the next step was delivering the benefit. But even this came with a dilemma: most students were under the age of 18 and legally required a guardian’s authorization to access the savings account created by the public bank.

One option was to require families to visit a physical bank branch to complete the authorization process. While this ensured legal and procedural rigor, it imposed a significant burden on families, especially those in rural areas, communities living along the riverbanks or with limited mobility.

The alternative was to allow digital authorization via the bank’s mobile app. This was more user-friendly and scalable, but raised concerns about identity verification. We feared that many guardians lacked the digital skills or trust in the system to complete the process independently. It was important to guarantee a friendly and user-centred way of providing the authorization.

We opted for the digital route, in line with the broader vision of Pé-de-Meia as a digitally enabled program. Yet the choice revealed a major bottleneck: the fact that hundreds of thousands of accounts remained inactive due to missing guardian authorizations. The issue exposed a key lesson: digitalization alone does not guarantee usability.

To address this, we developed a new strategy: using the WhatsApp messaging app to communicate directly with Pé-de-Meia students whose families had not completed the authorization process.  We were receiving over 100 daily requests — via emails, transparency portals, and other channels, under Brazil’s Access to Information Law — of students and families wanting to know about the payment of the installments and whether they were eligible for the program. WhatsApp offered a direct, low-friction way to communicate with beneficiaries and reinforce transparency at scale. However, as of of May 2025, more than 600,000 students had still not accessed their savings benefit

Actions and Reactions

Once the Ministry of Education decided to operationalize Pé-de-Meia through existing digital infrastructure, linking data from CadÚnico with enrollment and attendance records submitted by state education departments, there were profound ripple effects across Brazil’s federative system. While the logic of the policy seemed straightforward, translating it into action required adaptation across diverse and unequal territories.

The core operational rule that students must be enrolled and maintain at least 80% school attendance to receive their monthly incentives presented a serious challenge. In practice, many subnational education systems lacked reliable, centralized mechanisms to track this information. Attendance was often marked on paper, stored locally, or not monitored systematically. For these states and municipalities, Pé-de-Meia introduced a new operational urgency: without data interoperability, students could not receive their benefits.

The Ministry’s decision to require monthly data submission through the Sistema Gestão Presente (SGP), a federal educational data hub developed by the Federal University of Alagoas (UFAL), triggered mixed reactions. A few education secretariats had existing digital infrastructures and quickly integrated their systems with the national platform. But the majority found the requirement disruptive and burdensome, citing a lack of IT capacity, staff training, or the culture of collecting attendance data in a systematic way in schools.

In response, the federal team launched a series of onboarding sessions, capacity-building workshops, and technical support channels to guide local teams. Feedback loops were built into the system, including dashboards showing submission status by state and municipality. This helped identify bottlenecks and prioritize targeted support. The program also fostered a spirit of collaboration among state officials, who began exchanging practices and solutions in regular online forums.

One mid-sized state, with historically fragmented data collection systems, used Pé-de-Meia as a catalyst to transform its data practices. At the beginning of the program, school attendance was recorded manually and consolidated only at the end of each semester. With the need for real-time data, the state created a working group that brought together school administrators, local IT technicians, and education department staff. Within a few months, they had built a user-friendly attendance interface for school principals and created an API to automatically transfer data to the federal SGP platform, developed by the Federal University of Alagoas in partnership with the Ministry of Education. What began as a compliance task became an engine for digital innovation and long-term administrative capacity.

Reactions also came from students and families. While many welcomed the financial support, a large share of beneficiaries had limited digital access and did not initially understand how to monitor their eligibility or payments. The federal government had opted to use Jornada do Estudante (Student Journey app) as the main communication channel. However, early analytics revealed that nearly one-third of students had not logged into the app, highlighting a critical gap between design and user engagement.

To address this, several states launched outreach campaigns. Some sent SMS messages through schools; others held informational sessions in classrooms or created step-by-step guides on how to use the app. In certain municipalities, school staff helped students and families install the app and verify their information. These grassroots responses demonstrated how subnational actors adapted national digital tools to local realities, often playing a mediating role between the state and citizens.

Even within the federal government, the program reshaped internal workflows. Staff from different departments, such as education, social development, finance, and IT, had to coordinate daily, adapting to a new logic of interoperability. Policy success now depended not on isolated implementation, but on the fluid exchange of data, decisions, and responsibilities. This interministerial cooperation, although demanding, laid a foundation for more integrated digital governance in future policies.

Over time, what had begun as a new conditional transfer program took on a larger meaning. Pé-de-Meia encouraged subnational governments to build capacity, adapt digital tools, and reorganize processes. It created demand for public digital infrastructure where it didn’t previously exist and gave political visibility to backend systems that are often ignored, but essential.

These local adaptations illustrate how Pé-de-Meia became more than a conditional cash transfer,  helping to catalyze a broader shift toward digital-era public governance. More than just a funding mechanism, Pé-de-Meia redefined how the state interacts with citizens by making data, infrastructure, and coordination central to policy delivery. By positioning the federal government as an enabler of local innovation, the program embodies the principle of government as a platform, offering shared digital tools, standards, and data flows that empower subnational actors to design more responsive, citizen-centered solutions.

Further Reflections 

1. Governing Data in a Federative System

One of the most significant lessons from the Pé-de-Meia experience lies in the complexity of governing data across Brazil’s federative system. With 27 states and more than 5,500 municipalities, each with varying capacities for data management and digital infrastructure, creating a unified system to manage educational data was a daunting task. The Ministry of Education had to strike a delicate balance between ensuring standardization and respecting the autonomy of subnational entities.

This challenge revealed the importance of incentives and coordination mechanisms. Financial incentives, offered to students, had a spillover effect: they incentivized education departments to improve internal data systems, adopt new technologies, and participate in a federated platform. The decision to use the Present Management System (SGP), developed by a federal university, helped reduce asymmetries by offering a shared technical infrastructure that states could connect to. Still, this required intensive onboarding, continuous technical support, and political negotiation.

Moving forward, the Pé-de-Meia case raises important policy design questions: How can digital governance frameworks ensure reliability and interoperability without overburdening local institutions? What role should federal agencies play in building subnational digital capacity? And how can long-term data governance strategies be institutionalized so that they survive beyond the duration of a specific program? Pé-de-Meia offers valuable lessons on the need for shared standards, strong coordination, and a cooperative federalism mindset to govern data effectively at scale.

2. Digital Inclusion as a Prerequisite for Rights

Although Pé-de-Meia is grounded in digital innovation, its implementation exposed persistent gaps in digital inclusion. Despite the fact that the Jornada do Estudante app was designed to give beneficiaries transparency over their rights and payments, a significant portion of students never accessed it. The reasons varied: lack of internet access, limited digital literacy, outdated devices, or simply not knowing the app existed.

This gap is more than a communication problem. It reveals a structural vulnerability in policies mediated by digital infrastructure, because if beneficiaries cannot access the digital platform, they are effectively excluded from the policy itself. Or worse: they do not know the program exists. For a program designed to promote equity, this presents a contradiction that cannot be ignored in future public service design. In a digital government, access to the internet, knowledge of how to use applications, and confidence in public platforms are not luxuries; they are prerequisites for exercising one’s rights.

Addressing this requires more than technical solutions. It demands an integrated approach that includes hybrid solutions (combining digital and in-person support), sustained investment in connectivity, and partnerships with schools and education departments. Programs such as Pé-de-Meia must anticipate and plan for the diversity of their users' digital realities, because digital transformation cannot fulfill its inclusive promise unless access, usability, and literacy are treated as core pillars of policy design.

3. Government as a Platform: From Policy to Infrastructure

Perhaps one of the most powerful insights from Pé-de-Meia is how a policy initiative can evolve into an enabler of digital infrastructure. By layering services onto pre-existing systems, such as CadÚnico for socioeconomic targeting, the SGP for school data, and the Jornada do Estudante app for communication, the program demonstrated the power of government-as-a-platform. This approach allowed for rapid scaling.

Yet the experience also revealed the limitations of digital infrastructure when it operates in silos. The success of Pé-de-Meia hinged not just on having the right tools, but on connecting them through interoperability standards, shared protocols, and interministerial cooperation at an unprecedented level. The Ministry of Education and other government agencies had to operate as a network rather than a set of vertical silos, representing a fundamental shift in public administration culture. What began as an educational savings policy became a test case for horizontal digital governance.

Looking ahead, Pé-de-Meia offers a model for how governments can design digital infrastructure that is modular, adaptable, and citizen-centred. It shows that when infrastructure is built not just to support a single policy, but to enable future innovation, the public sector becomes more agile, inclusive, and responsive. This case also reinforces the importance of public digital infrastructure as a public good that must be continuously maintained, governed, and improved for the benefit of all users.

Conclusion 

The Pé-de-Meia program provides a compelling example of how a policy designed to address a deeply rooted social issue — high school dropout rates among low-income youth — can simultaneously become a vehicle for digital transformation in the public sector. While its primary goal was to reduce educational inequality through conditional financial incentives, the program’s success ultimately depended on the creation of a robust digital infrastructure, capable of integrating disparate data systems, engaging millions of users, and fostering collaboration across all levels of government. Several lessons emerge from this experience.

First, digital tools can dramatically enhance the effectiveness and reach of social policies, but only when they are paired with thoughtful design and strong governance. Pé-de-Meia leveraged pre-existing infrastructure such as CadÚnico and built new layers like the Present Management System to make real-time tracking of student attendance and enrollment data possible. However, these systems only worked because technical, political, and operational support mechanisms were put in place to ensure adoption across Brazil’s federative landscape.

Second, access to technology is not the same as digital inclusion. While the Student’s Path app expanded communication between government and students, a large portion of beneficiaries never accessed the platform. This exposed the gap between having a tool and knowing how to use it, which is a critical distinction for policymakers aiming to implement citizen-centred digital services.

Third, government-as-a-platform is not just a theory; it is a practical governance strategy. By treating data systems, applications, and coordination mechanisms as foundational infrastructure, the federal government enabled rapid policy deployment, encouraged subnational innovation, and laid the groundwork for future reforms. Pé-de-Meia illustrates how investing in shared digital infrastructure today can empower more responsive, equitable, and scalable public service delivery tomorrow, especially for those who need social policies the most.

Ultimately, Pé-de-Meia is more than a case of policy implementation; it is a case of transformation. It shows that with the right combination of vision, coordination, and infrastructure, governments can not only deliver benefits more effectively but also reshape how the state relates to its citizens: placing data, inclusion, and trust at the center of the digital-era public service.

References

Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47(2), 263–291.

Paes de Barros, R., Franco, S., Machado, L. M., & Rocha, G. (2021). Consequências da violação do direito à educação(1a ed). Autografia.

Pereira, V. A. (2016). From early childhood to high school: Three essays on the economics of education. Pontifícia Universidade Católica do Rio de Janeiro.

Pereira, V. (2022). Diagnóstico do abandono e da evasão escolar no Brasil. Instituto Mobilidade e Desenvolvimento Social. https://imdsbrasil.org/wp-content/uploads/2022/06/ImdsA001-2022-DiagnosticoAbandonoEvasaoEscolarBrasil.pdf