## Notes from 03 June 2025 [[2025-06-02|← Previous note]] ┃ [[2025-06-04|Next note →]] I was glad to see new developments coming out of São Paulo this week: the city is moving forward with what may be a significant attempt at a charter school–like model in Brazil. The municipal government [is set to launch a public call](https://g1.globo.com/sp/sao-paulo/noticia/2025/06/02/gestao-nunes-vai-iniciar-concessao-de-escolas-municipais-a-iniciativa-privada.ghtml) for nonprofit organizations to take over both pedagogical leadership and facilities management in three newly built public schools ([a pilot expected to expand to 50](https://www.metropoles.com/sao-paulo/como-prefeitura-pretende-implantar-modelo-privado-na-gestao-de-escolas) in a district that conservatively administers some [1,500 schools](https://eolgerenciamento.prefeitura.sp.gov.br/frmgerencial/NumerosCoordenadoria.aspx?Cod=000000)), under outcome-oriented agreements with clear and binding performance expectations. For those less familiar with charter schools: they are publicly funded institutions managed by non-state organizations (typically nonprofits) that operate with greater autonomy than traditional public schools while remaining accountable for student learning outcomes. They've been a key education reform tool in the United States for decades, with mixed results ranging from [strong success stories](https://credo.stanford.edu/reports/item/national-charter-school-study-iii/) to [not-that-good ones](https://ies.ed.gov/use-work/resource-library/report/evaluation-report/evaluation-charter-school-impacts-final-report). The model has also inspired adaptations elsewhere, such as _[academies](https://www.gov.uk/types-of-school/academies#:~:text=Academies%20have%20more%20control%20over,it%20must%20become%20an%20academy.)_ in the UK, _[collaboration schools](https://publicschoolpartnerships.co.za/)_ in South Africa, and _[colegios en concesión](http://administracionpublica.cide.edu/escuelas-de-concesion-en-colombia-una-apuesta-por-la-calidad-educativa/)_ in Colombia. As you might expect, the model takes many different shapes depending on context, regulation, and political will. This is a big deal for Brazilian education policy, though it's important to clarify what we're talking about here. This isn't about "school choice" as understood in the context of US charter schools. Parents won't get to choose to send their kids to these schools—enrollment will still follow São Paulo's standard residence-based allocation system. Instead, this is fundamentally about **operational efficiency** and testing new models of public service delivery. **The governance framework behind the experiment** In Brazil, public schools operate with very limited managerial autonomy. This stems partly from a well-intentioned commitment to equity: the belief that, to be fair, every school and every student should follow the same rules and receive the same resources, regardless of context. At the same time, there's genuine concern that giving too much discretion to individual principals could open doors to misuse of public funds. The net effect is a system safeguarded by multiple procedural controls (meant to ensure fairness and accountability) that can make even simple operational adjustments feel cumbersome. A similar logic underpins how we hire teachers. Our civil service exams for educators prioritize legal defensibility over classroom readiness: selection criteria are designed so that, in the event of a legal challenge, the government can clearly justify why candidate A was chosen over candidate B. That focus makes sense if your main goal is to avoid lawsuits, but it also means we often miss out on identifying and supporting people who have the pedagogy and passion to succeed in the classroom. Seen from this perspective, the system's procedural strictness is less about obstructionism and more about protecting public resources and ensuring a level playing field—yet it can also erect barriers to innovation. Given these constraints, many policymakers approach alternative governance proposals with caution, not because they distrust change but because they genuinely fear unintended consequences: a more autonomous school might outperform its neighbors, creating pockets of inequality that clash with our unified public system. The question worth asking, however, is whether allowing some managed variation might actually strengthen the overall ecosystem. Could a more adaptive, flexible school network—one in which successful practices spread rather than remain siloed—ultimately deliver more equitable outcomes? This brings us to Brazil's _Organização Social_ (OS) model, our adaptation of the UK's quasi-autonomous non-governmental organizations, or quangos. Instead of converting public entities into semi-public ones, we contract nonprofit civil society organizations to manage public services for a defined period while keeping infrastructure under government ownership. We've seen it work in healthcare: hospitals and clinics run by _OS_ partners have been able to respond more quickly to local needs, streamline procurement, and link budgeting directly to outcomes ([OECD, 2021, p. 99](https://www.oecd.org/en/publications/2021/12/oecd-reviews-of-health-systems-brazil-2021_6797af6a.html)). The result has been greater agility and, in many cases, improved service quality—while infrastructure and oversight remain public. Could the same approach help public education? By contracting experienced nonprofits to run pedagogy and facilities, we might preserve core equity safeguards—like a common curriculum, participation in public evaluations, inclusive admissions, shared student identity markers (such as the standard public school uniform), and a strict no-fee policy—while granting schools the flexibility to respond to their own challenges. It's not about abandoning fairness; it's about experimenting with a model that has already delivered positive results elsewhere in the Brazilian public sector. In doing so, we can test whether managed autonomy can coexist with our commitment to a unified, publicly accountable school network—and learn what works, and for whom, without risking wholesale fragmentation. **Why this experiment matters** Some reports say the city will even mandate the same school supplies for every student, a clear signal that officials worry about fragmentation and inequity. I'm skeptical: yes, a shared curriculum and identical materials enable joint evaluations and comparability, but the real promise here is giving schools genuine pedagogical freedom. Is sticking to uniform textbooks and supplies enough to prevent divergence, or does it risk turning autonomy into mere window dressing? This tension between centralized coherence and local innovation is exactly what to watch. Brazil's education system has long shied away from bold service-delivery experiments, even as our health sector embraces public-private collaboration. Yet US charter school research reminds us that new governance alone is not a silver bullet: implementation can make or break success. That's precisely why we need to pilot this model—to see how local context shapes outcomes, rather than assuming failure or success by proxy. Of course, the real test isn't the contract itself but the city's ability to become a savvy commissioner: selecting partners wisely, writing clear contracts, and monitoring performance to encourage improvement, not just compliance. How do you evaluate proposals when you've never done it before? How do you monitor progress in ways that drive better teaching and learning, rather than merely ticking boxes? Critics will point out that building such institutional muscle is a monumental task, and they're right. But if we don't learn how to contract effectively, we'll never know whether managed autonomy can strengthen our public network rather than undermine it. Ultimately, this pilot isn't about proving a point; [it's about gathering evidence](https://x.com/snoble/status/1417254206088519691). Will a controlled experiment spark improvement without balkanizing our system? Only rigorous evaluation and honest dialogue across political lines will tell. In the meantime, philanthropic organizations and local governments must collaborate on multiple pilots, share lessons, and ensure each experiment brings us closer to a more responsive, equitable public school network. **The bottom line** This São Paulo pilot deserves careful monitoring and rigorous evaluation. The goal isn't to privatize education or create school choice, but to harness external capacity and innovation to improve public service delivery. If successful, it could provide a roadmap for introducing beneficial management innovation into Brazilian public education while maintaining our commitment to universal, equitable access. If it doesn't work, we'll learn valuable lessons about implementation challenges in our specific context. Either way, it's about time we started experimenting. Brazilian education policy has been too risk-averse for too long, and our students deserve better than perpetual incrementalism.