Open education policies aren't enough to fix global schooling gaps

The results show that, on average, OER policy adoption has not yet led to statistically robust improvements in these national indicators. Countries that adopted OER frameworks did not experience large or immediate increases in completion rates compared with non-adopting countries. Similarly, reductions in out-of-school rates associated with OER adoption were modest and statistically imprecise.

New global evidence suggests that simply adopting open education policies does not automatically translate into measurable gains in access or completion, at least not in the short term. Instead, a country’s broader digital and institutional capacity appears to play a far more decisive role in shaping whether these policies deliver results.

A newly published study, Bridging Digital Readiness and Educational Inclusion: The Causal Impact of OER Policies on SDG4 Outcomes, appearing in the journal Sustainability, evaluates the real-world impact of national open educational resource policies across 187 countries between 2015 and 2024. The research assesses how these policies interact with artificial intelligence readiness, digital infrastructure, and governance capacity to influence progress toward Sustainable Development Goal 4.

OER policies alone show limited short-term impact on SDG4 indicators

Open educational resources have become a cornerstone of international education reform efforts over the past decade. Designed to provide free, adaptable learning materials, OER policies are intended to reduce textbook costs, expand access, and support inclusive education systems. Between 2015 and 2024, more than thirty countries introduced formal national OER frameworks through legislation, ministerial mandates, or comprehensive digital education strategies.

The study evaluates whether these policy moves translated into measurable outcomes using two key SDG4 indicators: primary school completion rates and out-of-school rates for primary-aged children. To isolate policy effects, the authors applied fixed-effects models and difference-in-differences analysis, allowing them to control for country-specific characteristics and global shocks affecting education systems over time.

The results show that, on average, OER policy adoption has not yet led to statistically robust improvements in these national indicators. Countries that adopted OER frameworks did not experience large or immediate increases in completion rates compared with non-adopting countries. Similarly, reductions in out-of-school rates associated with OER adoption were modest and statistically imprecise.

This pattern does not imply that OER policies are ineffective. Rather, the authors emphasize that most national frameworks are still in early implementation stages. The majority of policies were introduced between 2019 and 2022, leaving only two to five years of post-adoption data available for analysis. Education systems, by nature, change slowly. Primary education cycles often span six to eight years, meaning that many students captured in recent completion data began schooling before OER policies were in place or experienced only partial exposure.

The study also highlights structural data constraints that limit the ability to detect early effects. Because many countries adopted policies close to the end of the observation period, there is insufficient within-country variation to isolate policy impacts cleanly using fixed-effects methods. While difference-in-differences models suggest a small positive trend for OER-adopting countries after 2020, the uncertainty around these estimates remains large.

The findings suggest that expectations of rapid, system-wide transformation from OER policies may be unrealistic. Early-stage implementation, gradual scaling, and the long timelines of educational reform mean that measurable national effects are unlikely to appear immediately.

AI readiness emerges as a stronger predictor of education outcomes

While the direct impact of OER policies remains limited in the short term, the study identifies artificial intelligence readiness as a far more consistent and influential factor shaping education outcomes. AI readiness, as measured by the Oxford Insights Government AI Readiness Index, captures a country’s capacity across digital infrastructure, governance quality, technical workforce, and innovation ecosystems.

Across multiple model specifications, higher AI readiness is associated with higher primary completion rates and lower out-of-school rates. A ten-point increase in the readiness index corresponds to roughly a half-percentage-point increase in completion rates and a notable reduction in the share of children excluded from schooling. These associations persist even after controlling for internet use and broadband penetration.

The findings suggest that digital readiness reflects more than connectivity alone. Countries scoring high on AI readiness tend to have stronger regulatory frameworks, better data governance, more skilled educators and technical staff, and greater institutional capacity to deploy and manage digital tools effectively. These conditions support not only advanced technologies but also the practical implementation of education reforms.

Importantly, the study finds that AI readiness also moderates the effectiveness of OER policies. Interaction analyses show that open education frameworks are more likely to be associated with positive outcomes in countries that already possess high digital and institutional capacity. In contrast, low-income and low-readiness countries show little evidence of short-term benefits from OER adoption.

This pattern underscores a central conclusion of the research: open educational resources function as complementary interventions rather than standalone solutions. Without reliable infrastructure, trained teachers, clear governance structures, and innovation support, OER policies risk remaining aspirational documents rather than engines of change.

The authors caution against one-size-fits-all policy recommendations. While digitally advanced countries may be well positioned to leverage OER policies quickly, others face more fundamental constraints that must be addressed first. In settings where internet access is uneven, teacher training is limited, or institutional coordination is weak, the potential benefits of open resources are harder to realize.

Policy implications point to sequencing, capacity building, and patience

The study’s findings carry significant impact for governments, development agencies, and international organizations seeking to accelerate progress toward SDG4. Key among them is the need to rethink how and when OER policies are deployed.

Rather than treating policy adoption as an endpoint, the research emphasizes the importance of sequencing reforms in line with national capacity. In countries with strong digital readiness, OER frameworks can act as accelerators, building on existing infrastructure and skills to expand access and reduce costs. In lower-capacity contexts, however, policy adoption without parallel investment in connectivity, governance, and human capital may yield limited returns.

This does not mean that low-readiness countries should abandon OER strategies. Formal policies can still play a valuable signaling role, establishing standards, encouraging coordination, and attracting future investment. However, expectations should be calibrated to reflect longer timelines and the need for sustained capacity building.

For international development partners, the findings argue for integrated support packages rather than isolated interventions. Investments in broadband expansion, teacher digital training, content development, and governance reform are likely to reinforce one another. Supporting OER policies without addressing these foundations risks diluting impact and reinforcing inequalities between countries at different stages of digital development.

The study also contributes to ongoing debates about evidence-based education reform. By focusing on early-stage policy impacts, it highlights the dangers of judging complex interventions too quickly. Modest or statistically uncertain effects in the first few years do not necessarily indicate failure. Many education reforms show limited initial results before gains accumulate as implementation matures and systems adapt.

Furthermore, the authors call for continued monitoring and longer-term analysis. As OER policies age and more cohorts of students experience full exposure to open resources, clearer patterns may emerge. Data extending toward 2030 will be critical for determining whether early signals strengthen into sustained improvements or plateau over time.